{
  "schema": 1,
  "signature": "3da4d212825e57cd519f2212854fc4e9af108bd602fc43b2784a4ba5d89cc276",
  "environment": {
    "measured_at": "2026-10-12T16:46:02.648471+08:00",
    "system": "Linux",
    "kernel": "7.0.0-34-generic",
    "machine": "x86_64",
    "cpu": {
      "lscpu": [
        {
          "field": "Architecture:",
          "data": "x86_64"
        },
        {
          "field": "CPU op-mode(s):",
          "data": "32-bit, 64-bit"
        },
        {
          "field": "Address sizes:",
          "data": "46 bits physical, 48 bits virtual"
        },
        {
          "field": "Byte Order:",
          "data": "Little Endian"
        },
        {
          "field": "CPU(s):",
          "data": "16"
        },
        {
          "field": "On-line CPU(s) list:",
          "data": "0-15"
        },
        {
          "field": "Vendor ID:",
          "data": "GenuineIntel"
        },
        {
          "field": "Model name:",
          "data": "12th Gen Intel(R) Core(TM) i7-1270P"
        },
        {
          "field": "CPU family:",
          "data": "6"
        },
        {
          "field": "Model:",
          "data": "154"
        },
        {
          "field": "Thread(s) per core:",
          "data": "2"
        },
        {
          "field": "Core(s) per socket:",
          "data": "12"
        },
        {
          "field": "Socket(s):",
          "data": "1"
        },
        {
          "field": "Stepping:",
          "data": "3"
        },
        {
          "field": "CPU(s) scaling MHz:",
          "data": "70%"
        },
        {
          "field": "CPU max MHz:",
          "data": "4800.0000"
        },
        {
          "field": "CPU min MHz:",
          "data": "400.0000"
        },
        {
          "field": "BogoMIPS:",
          "data": "4992.00"
        },
        {
          "field": "Flags:",
          "data": "fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk avx_vnni dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req hfi vnmi umip pku ospke waitpkg gfni vaes vpclmulqdq rdpid movdiri movdir64b fsrm md_clear serialize pconfig arch_lbr ibt flush_l1d arch_capabilities"
        },
        {
          "field": "Virtualization:",
          "data": "VT-x"
        },
        {
          "field": "L1d cache:",
          "data": "448 KiB (12 instances)"
        },
        {
          "field": "L1i cache:",
          "data": "640 KiB (12 instances)"
        },
        {
          "field": "L2 cache:",
          "data": "9 MiB (6 instances)"
        },
        {
          "field": "L3 cache:",
          "data": "18 MiB (1 instance)"
        },
        {
          "field": "NUMA node(s):",
          "data": "1"
        },
        {
          "field": "NUMA node0 CPU(s):",
          "data": "0-15"
        },
        {
          "field": "Vulnerability Gather data sampling:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Ghostwrite:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Indirect target selection:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Itlb multihit:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability L1tf:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Mds:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Meltdown:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Mmio stale data:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Old microcode:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Reg file data sampling:",
          "data": "Mitigation; Clear Register File"
        },
        {
          "field": "Vulnerability Retbleed:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Spec rstack overflow:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Spec store bypass:",
          "data": "Mitigation; Speculative Store Bypass disabled via prctl"
        },
        {
          "field": "Vulnerability Spectre v1:",
          "data": "Mitigation; usercopy/swapgs barriers and __user pointer sanitization"
        },
        {
          "field": "Vulnerability Spectre v2:",
          "data": "Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S"
        },
        {
          "field": "Vulnerability Srbds:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Tsa:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Tsx async abort:",
          "data": "Not affected"
        },
        {
          "field": "Vulnerability Vmscape:",
          "data": "Mitigation; IBPB before exit to userspace"
        }
      ]
    },
    "cpu_topology": "# The following is the parsable format, which can be fed to other\n# programs. Each different item in every column has an unique ID\n# starting usually from zero.\n# CPU,Core,Socket,Maxmhz\n0,0,0,4800.0000\n1,0,0,4800.0000\n2,1,0,4800.0000\n3,1,0,4800.0000\n4,2,0,4800.0000\n5,2,0,4800.0000\n6,3,0,4800.0000\n7,3,0,4800.0000\n8,4,0,3500.0000\n9,5,0,3500.0000\n10,6,0,3500.0000\n11,7,0,3500.0000\n12,8,0,3500.0000\n13,9,0,3500.0000\n14,10,0,3500.0000\n15,11,0,3500.0000\n",
    "affinity": "0,2,4,6",
    "monitor_affinity": "8,9,10,11,12,13,14,15",
    "memory_total_kib": 32409724,
    "loadavg_start": [
      9.50537109375,
      9.87548828125,
      10.072265625
    ],
    "cgroup_cpu_max": null,
    "cgroup_cpuset": "0-15",
    "cgroup_membership": "0::/user.slice/user-1000.slice/session-5.scope",
    "governor": "powersave",
    "seed": 20261012,
    "fresh_processes": 3,
    "warm_samples_per_process": 5,
    "profile": "baseline",
    "stress_seconds": 0,
    "memory_guard_tree_rss_mib": 4096,
    "memory_guard_interval_ms": 1000,
    "memory_passes_per_case": 1,
    "pss_interval_ms": 10,
    "memory_final_hold_ms": 100,
    "timing_memory_separate": true,
    "cold_cache": false,
    "thread_env": {
      "OPENBLAS_NUM_THREADS": "1",
      "MKL_NUM_THREADS": "1",
      "OMP_DYNAMIC": "FALSE",
      "NUMBA_THREADING_LAYER": "omp"
    },
    "sources_sha256": {
      "native/build.sh": "ca8699ae1d635f79411e457d1fa3b75dbe4bf5e0464a536e1a779ccacf120fce",
      "native/parallel.cpp": "e85e7462bc88169b8d7ebe3e8468d464f12f25e9c7a99860812012fb81bf5ab8",
      "native/rust/Cargo.lock": "96bdab99750bea2a8cb3439bbbbb025e4ba6128784686b3674ce1c61daedefe7",
      "native/rust/Cargo.toml": "31c81f7598d66220514cca81e57776ec7d82d33c67de1709a3c91c2bd7bee5ea",
      "native/rust/src/main.rs": "b3374aafb21d65d900689c3d5bd57572e012954d44ed1fc8ea5debfd0129ff3d",
      "python/bench.py": "c7ae942e726a02b2fec53f4fe21a1cb7f5c1d8efadca6e97cefe3365cf148ecf",
      "python/requirements.txt": "86a728929b42a7aeacf5bfaaa91e09812913a5da0a32a354dbed0658e3f88f48",
      "reference.py": "eae23aefdb36731276e633628e8666167b71041b4874d65b71b48d71cd87ef1a",
      "run.py": "2814722d250609807ec004ce9430f65c6380a24f7277d8e83aae31d0cefce2e6"
    },
    "executables": {
      "cpp": {
        "sha256": "11cec31bcb15a7500d96b6f13082f1f7965fb5c7f497a2af20d375d9d3ce729f",
        "name": "parallel-cpp"
      },
      "rust": {
        "sha256": "eb6c38a55d37f2a9bc70778fb343e040f9a65851ecbf5e7760eeadc5865849aa",
        "name": "parallel-rust"
      },
      "python": {
        "sha256": "e50d468e8b0adfb05733f5b87b3cff34829c4a8c1aea50c865aa8bdfe4bb150f",
        "name": "python"
      },
      "python_script": {
        "sha256": "c7ae942e726a02b2fec53f4fe21a1cb7f5c1d8efadca6e97cefe3365cf148ecf",
        "name": "bench.py"
      }
    },
    "python_runtime_versions": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]\n2.2.6 1.15.3 0.61.2 0.44.0\n",
    "loadavg_end": [
      10.8359375,
      10.5595703125,
      10.32177734375
    ]
  },
  "references": {
    "dense": {
      "checksum": "41338",
      "output_hash": "0db0f71d8f68135f",
      "logical_data_bytes": 33587200,
      "stored_entries": 4194304,
      "nonzero_entries": 4152776,
      "max_abs_q": 38803,
      "output_scale": 1024
    },
    "sparse-uniform": {
      "checksum": "7278",
      "output_hash": "18cc5ed244ef97d9",
      "logical_data_bytes": 25952264,
      "stored_entries": 2097152,
      "nonzero_entries": 2076388,
      "max_abs_q": 5430,
      "output_scale": 1024
    },
    "sparse-skew": {
      "checksum": "538",
      "output_hash": "42c390bd74645f49",
      "logical_data_bytes": 25952264,
      "stored_entries": 2097152,
      "nonzero_entries": 2076388,
      "max_abs_q": 7459,
      "output_scale": 1024
    }
  },
  "records": [
    {
      "case": {
        "id": "python:thread/dense/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1287.9668519599363,
        "first_ms": 440.57610095478594,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2193.8878069631755,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          449.35067906044424,
          433.189157047309,
          433.73095500282943,
          436.5549370413646,
          438.3698560995981
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3996.484039002098,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 517.9626710014418,
        "first_ms": 992.8628029301763,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.320219018496573,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.279211978428066,
          8.536998997442424,
          2.921594073995948,
          6.816543987952173,
          2.8762080473825336
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1800.7276098942384,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 800.3887579543516,
        "first_ms": 444.9255019426346,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1168.0217540124431,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          212.3701290693134,
          202.59262097533792,
          259.691154002212,
          204.83576005790383,
          227.9663230292499
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2523.575723054819,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 542.561762034893,
        "first_ms": 516.200160025619,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.92017107084394,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.073963056318462,
          2.733201952651143,
          2.0173259545117617,
          2.9357189778238535,
          2.5311369681730866
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1356.7708110203966,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.823758,
        "first_ms": 2.568394,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.779631,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.395305,
          4.935107,
          2.326351,
          2.991595,
          2.61867
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 44.812121079303324,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.207085,
        "first_ms": 2.790477,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.409742,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.44302,
          2.456822,
          1.979186,
          1.923095,
          3.322931
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 41.86070105060935,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/dense/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 990.5654700705782,
        "first_ms": 327.26915495004505,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1709.0242009144276,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          328.0389659339562,
          345.1197649119422,
          351.35353601071984,
          344.94554495904595,
          335.13414103072137
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3060.000687954016,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.599587,
        "first_ms": 2.581156,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.86047,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.941702,
          2.97445,
          1.96461,
          2.697321,
          1.994488
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30120,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 41.10894200857729,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 841.1833540303633,
        "first_ms": 535.0316079566255,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1392.2498050378636,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          302.8840370243415,
          260.3723850334063,
          241.53898702934384,
          254.55736310686916,
          256.7003429867327
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2895.2509270748124,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 925.1572909997776,
        "first_ms": 13.256250997073948,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 131.90061994828284,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.676489995792508,
          6.362528074532747,
          5.6359609588980675,
          5.5318999802693725,
          10.46097301878035
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1266.484974999912,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 40.951848,
        "first_ms": 7.380852,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 33.851759,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.37275,
          6.749779,
          4.960401,
          6.539766,
          9.98002
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 91.00515791215003,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-skew/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 841.3866889895871,
        "first_ms": 209.92281893268228,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1117.0940559823066,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          208.2193500827998,
          207.231329055503,
          212.45986700523645,
          209.55645700450987,
          207.4480860028416
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2226.8469349946827,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 834.7186599858105,
        "first_ms": 292.0048249652609,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1253.4030419774354,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          235.2206140058115,
          236.43123102374375,
          236.14097794052213,
          235.52718095015734,
          239.22879889141768
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2456.2569139525294,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 41.645706,
        "first_ms": 6.528808,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 34.16553,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.938356,
          6.85044,
          5.959409,
          6.867758,
          8.073309
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 93.30941701773554,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.318438,
        "first_ms": 4.368038,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.265855,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.261258,
          4.162209,
          3.523638,
          3.973032,
          4.321049
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36976,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 62.707356992177665,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 26.559987,
        "first_ms": 4.290387,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 23.518971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.095793,
          4.058438,
          5.383302,
          3.518008,
          4.436338
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36936,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 61.51452101767063,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1091.3191509898752,
        "first_ms": 565.8303499221802,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1719.1152269952,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          310.3109629591927,
          363.26718202326447,
          301.1354700429365,
          370.16540800686926,
          371.4565068949014
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3469.3742389790714,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.606489,
        "first_ms": 2.886385,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.501407,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.451363,
          3.388572,
          2.72677,
          2.349196,
          2.073105
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 41.57973197288811,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.652066,
        "first_ms": 3.371399,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.494593,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.005091,
          2.481766,
          2.530692,
          2.184873,
          2.00667
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.321651915088296,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.32193,
        "first_ms": 2.783669,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.677147,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.557246,
          2.431051,
          2.381459,
          3.13927,
          2.679189
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 42.02974995132536,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/dense/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.305699,
        "first_ms": 4.026939,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.386722,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.053144,
          6.294939,
          3.373161,
          5.374983,
          3.257735
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 50.97735405433923,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 572.7542579406872,
        "first_ms": 525.7918660063297,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 95.17757990397513,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.430350985378027,
          2.463288023136556,
          2.6100960094481707,
          2.57001502905041,
          3.1736460514366627
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1410.7307828962803,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 834.3763339798898,
        "first_ms": 208.49967503454536,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1101.3995440443978,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          207.48523296788335,
          205.1974789937958,
          206.1271530110389,
          205.15290088951588,
          207.8798779984936
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2194.2888220073655,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.062992,
        "first_ms": 2.582687,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.808509,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.902593,
          2.561549,
          2.561896,
          2.414389,
          2.877931
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 38.120048004202545,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 830.6983310030773,
        "first_ms": 379.5663519995287,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 771.1668729316443,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          163.55662094429135,
          132.6981910970062,
          129.0302569977939,
          124.87670290283859,
          126.36769295204431
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2079.263301915489,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 784.399012918584,
        "first_ms": 263.15496605820954,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1227.3347359150648,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          216.1081830272451,
          194.42607800010592,
          206.03044400922954,
          258.1402739742771,
          258.3035769639537
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2395.1002929825336,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.33868,
        "first_ms": 3.904352,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.028604,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.878546,
          5.213468,
          3.286553,
          4.395623,
          3.219862
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 48.67538693360984,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.971218,
        "first_ms": 2.812424,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.951768,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.980886,
          3.688966,
          2.488766,
          2.184681,
          4.289958
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30200,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 49.32512901723385,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 36.781106,
        "first_ms": 2.909123,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.985682,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.792391,
          2.091997,
          1.930989,
          6.945264,
          1.945089
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 68.3719830121845,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 31.45383,
        "first_ms": 4.459189,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 20.872651,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.261999,
          4.738065,
          3.236642,
          3.94299,
          5.659237
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 65.39261003490537,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 41.625856,
        "first_ms": 2.100078,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.921664,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.758333,
          2.355584,
          4.420653,
          5.186846,
          1.923454
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30224,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 80.12253406923264,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 32.013399,
        "first_ms": 8.470848,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 42.76875,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.815882,
          7.119204,
          10.872655,
          10.713562,
          7.211421
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 91.42737893853337,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1328.6105829756707,
        "first_ms": 385.0524789886549,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1259.5905689522624,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          302.70623101387173,
          229.1884000878781,
          246.12603394780308,
          238.36051707621664,
          240.27377599850297
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3027.9610590077937,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1049.8613560339436,
        "first_ms": 365.05472799763083,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1294.162357924506,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          260.8238860266283,
          266.8942059390247,
          259.3750739470124,
          249.90616191644222,
          254.39303100574762
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2783.741423045285,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 40.402474,
        "first_ms": 6.399296,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 20.244335,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.219803,
          4.781458,
          8.702245,
          2.008408,
          2.120285
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 76.99722703546286,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 598.4302449505776,
        "first_ms": 2.7730880538001657,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 86.51055896189064,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.085738000459969,
          3.0943240271881223,
          2.2565360413864255,
          2.3360539926216006,
          3.1164020765572786
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 788.7365630595013,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1024.4837930658832,
        "first_ms": 440.8363300608471,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1645.9168980363756,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          354.1240270715207,
          339.2205099808052,
          304.15881401859224,
          314.68542595393956,
          329.7106600366533
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3194.493955001235,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.526583,
        "first_ms": 3.554371,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.67724,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.857221,
          3.225735,
          3.058268,
          4.48428,
          3.02996
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37652,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 51.63753195665777,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 568.6715020565316,
        "first_ms": 529.5623369747773,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.80872004758567,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.1526120146736503,
          2.629810944199562,
          2.302483073435724,
          2.163612050935626,
          2.144285012036562
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1387.5327750574797,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 807.9197860788554,
        "first_ms": 501.9086350221187,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1502.837879001163,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          297.59279591962695,
          279.80662893969566,
          282.4394170893356,
          295.8011949667707,
          273.14355003181845
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2884.4877049559727,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.205036,
        "first_ms": 2.797385,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.30478,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.773244,
          3.000686,
          3.493956,
          2.494493,
          2.514241
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37024,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 49.96467393357307,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.094705,
        "first_ms": 3.286692,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.256374,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.508172,
          2.003673,
          2.010365,
          1.997516,
          3.471737
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.03234603442252,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 125.79197401646525,
        "first_ms": 3.8675640244036913,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.157579983584583,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.3779629739001393,
          2.769788959994912,
          3.553098998963833,
          2.8984000673517585,
          1.9063029903918505
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 194.61563194636256,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.326865,
        "first_ms": 3.143504,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.958971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.17521,
          2.984984,
          2.954317,
          4.502902,
          3.3228
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37096,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 49.265378969721496,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 655.6121150497347,
        "first_ms": 521.0026800632477,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2352.0924350013956,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          469.70460494048893,
          463.58202607370913,
          465.322058997117,
          469.08830606844276,
          481.71718802768737
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3603.6130680004135,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 35.498224,
        "first_ms": 3.353867,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.96212,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.339167,
          3.544594,
          4.527015,
          2.237321,
          5.050036
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 70.49792201723903,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 997.6554920431226,
        "first_ms": 323.7717649899423,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 944.4444669643417,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          181.39795598108321,
          196.20269897859544,
          195.59253996703774,
          192.39557697437704,
          174.6403370052576
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2343.6437679920346,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 276.07119595631957,
        "first_ms": 537.4324340373278,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.713683989830315,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.721665940247476,
          4.291919060051441,
          3.510736976750195,
          5.269601009786129,
          3.4836989361792803
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1020.9492580033839,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.589034,
        "first_ms": 2.596087,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.387667,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.386784,
          4.002592,
          2.335253,
          2.405621,
          3.779018
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 42.81719995196909,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 828.1148569658399,
        "first_ms": 210.53811395540833,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1252.5044600479305,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          261.38196501415223,
          285.46207305043936,
          207.47873303480446,
          209.01513996068388,
          206.87277999240905
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2339.3406359246,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-uniform/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 813.3335050661117,
        "first_ms": 209.40250402782112,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1114.3425509799272,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          209.25709698349237,
          206.23234403319657,
          210.3280610172078,
          209.10032698884606,
          210.15124500263482
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2182.468288927339,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.100439,
        "first_ms": 2.774032,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.920585,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.93695,
          2.109111,
          2.043865,
          2.642339,
          2.907724
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.709890054538846,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-skew/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.214942,
        "first_ms": 2.531091,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.712785,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.036449,
          2.591479,
          1.970618,
          2.764303,
          2.088078
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.455279940739274,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.062928,
        "first_ms": 2.831088,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.891126,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.054686,
          2.903902,
          2.091994,
          2.3988,
          2.16304
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30104,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.82919703796506,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.816031,
        "first_ms": 3.351391,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.731703,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.393457,
          2.811791,
          3.269088,
          2.400829,
          2.3611
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 39.53541489318013,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.221695,
        "first_ms": 3.489325,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.2192,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.02234,
          4.018938,
          3.038431,
          2.969172,
          4.152512
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37104,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 50.01314589753747,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 820.4807809088379,
        "first_ms": 211.49809891358018,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1089.7154979174957,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          202.08023593295366,
          205.0095689482987,
          207.62760599609464,
          204.37676599249244,
          204.0412169881165
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2167.8667940432206,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 17.820313,
        "first_ms": 4.010774,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.972082,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.102214,
          4.352911,
          3.026471,
          3.166826,
          4.293505
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 46.924202935770154,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 534.7539769718423,
        "first_ms": 3.5694669932127,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 87.58582395967096,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.7007380267605186,
          2.175802015699446,
          2.2438809974119067,
          2.1424139849841595,
          2.751133986748755
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 711.1910019302741,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.265367,
        "first_ms": 2.937879,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.365371,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.0803,
          3.022299,
          3.296604,
          2.034916,
          1.640973
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.883673005737364,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1022.1713660284877,
        "first_ms": 562.2132769785821,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2461.8828530656174,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          485.81622599158436,
          520.7985689630732,
          484.05319498851895,
          478.0322159640491,
          488.7827590573579
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 4103.6585019901395,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.812838,
        "first_ms": 4.408347,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.057806,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.305295,
          3.861293,
          3.946595,
          3.25047,
          3.671621
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37652,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 53.44231997150928,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 24.351409,
        "first_ms": 3.704892,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.450061,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.746289,
          3.21387,
          3.283384,
          4.889732,
          3.293874
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37784,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 57.483099051751196,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-skew/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 39.303939,
        "first_ms": 3.043605,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.775995,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.515872,
          2.533084,
          4.202533,
          2.488315,
          2.501767
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 63.02131398115307,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.804273,
        "first_ms": 3.474972,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.166778,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.381797,
          3.622397,
          3.358109,
          4.470873,
          3.296307
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 58.368397993035614,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 608.1609459361061,
        "first_ms": 336.6569150239229,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.37388302385807,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.001799992285669,
          3.3614039421081543,
          2.4614569265395403,
          1.956893946044147,
          1.976636005565524
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1242.694205022417,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 882.9461509594694,
        "first_ms": 378.5831310087815,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1279.2645430890843,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          242.33214696869254,
          239.49277598876506,
          243.38602309580892,
          246.46818300243467,
          236.5935390116647
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2639.999584062025,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 581.8612179718912,
        "first_ms": 564.8951280163601,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.65897998213768,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.4171960540115833,
          2.133582951501012,
          2.720469026826322,
          2.114937989972532,
          2.904004999436438
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1458.253973047249,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.468183,
        "first_ms": 3.153717,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.161078,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.22138,
          2.968359,
          1.96417,
          1.968915,
          2.747549
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 41.74522194080055,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-skew/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 571.8005510279909,
        "first_ms": 3.284853999502957,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 94.64135998860002,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.002550966106355,
          2.154362970031798,
          2.653567004017532,
          2.6885049883276224,
          2.9729759553447366
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 761.7268890608102,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 132.09274294786155,
        "first_ms": 2.4836710654199123,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.468216992914677,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.219522932544351,
          2.2384029580280185,
          1.9990981090813875,
          2.9774309368804097,
          2.027184935286641
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 204.6296939952299,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/dense/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 132.26321898400784,
        "first_ms": 4.309103009290993,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.249626060016453,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          1.9885170040652156,
          1.9016950391232967,
          3.6569139920175076,
          1.9711790373548865,
          1.9938969053328037
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 204.0916329715401,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 830.7105370331556,
        "first_ms": 212.67444593831897,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1123.5215530032292,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          209.99859203584492,
          213.16046407446265,
          210.51585406530648,
          209.6565399551764,
          212.9160879412666
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2215.2193599613383,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.82696,
        "first_ms": 3.52693,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.500553,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.812838,
          2.601779,
          3.438846,
          2.164096,
          2.951153
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 45.36744102369994,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 561.7035940522328,
        "first_ms": 3.815196920186281,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.33029598928988,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.981032990850508,
          3.1205270206555724,
          3.3087250776588917,
          2.1830840269103646,
          2.8252010233700275
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 749.4946729857475,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.355285,
        "first_ms": 2.51294,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 27.738907,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.183137,
          5.050932,
          5.901932,
          6.820587,
          7.421495
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 61.306533054448664,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 41.914401,
        "first_ms": 3.201923,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.365858,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.401213,
          3.225088,
          3.109995,
          4.416333,
          3.193582
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36948,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 72.147605009377,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.904589,
        "first_ms": 2.829808,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.088246,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.946701,
          2.035577,
          3.005049,
          2.091393,
          2.727786
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30208,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.53300395794213,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/dense/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 286.34817304555327,
        "first_ms": 337.81525294762105,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 30.17443895805627,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.1274619763717055,
          3.7540659541264176,
          6.925056921318173,
          5.083882017061114,
          4.217181005515158
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 846.0385479265824,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 20.343838,
        "first_ms": 4.747382,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.007567,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.97016,
          3.934241,
          3.295672,
          3.827737,
          3.946368
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 49.39790000207722,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 824.247932061553,
        "first_ms": 412.1586650144309,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1473.46677898895,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          289.4783599767834,
          285.677285050042,
          266.7606499744579,
          278.7270799744874,
          266.9619229855016
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2785.0264291046187,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 851.1490810196847,
        "first_ms": 420.491985976696,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1750.6458510179073,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          330.2291949512437,
          346.9733780948445,
          361.4604000467807,
          312.7934290096164,
          322.70791998598725
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3117.8693340625614,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.389661,
        "first_ms": 3.115674,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.862831,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.678003,
          3.772105,
          2.699649,
          2.890702,
          3.496789
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30112,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 45.42700503952801,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.675537,
        "first_ms": 3.944765,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.935194,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.468171,
          3.163464,
          4.073078,
          3.078034,
          3.628041
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 48.766660038381815,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 22.278145,
        "first_ms": 3.226782,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.789314,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.217639,
          3.613169,
          2.758379,
          4.199035,
          4.60954
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 53.00519394222647,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 22.017193,
        "first_ms": 4.683189,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.570346,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.773494,
          3.481197,
          3.555768,
          2.905671,
          3.434701
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 51.35295097716153,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 721.8519389862195,
        "first_ms": 3.5051261074841022,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 103.47972007002681,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          7.170509081333876,
          2.879404928535223,
          3.794496995396912,
          3.181873937137425,
          4.945135908201337
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 993.1046499405056,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.744833,
        "first_ms": 3.587126,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.912152,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.93071,
          3.401407,
          3.337705,
          2.711105,
          3.231658
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 45.877494965679944,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.794625,
        "first_ms": 3.810054,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.856063,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.135203,
          3.255342,
          3.750148,
          3.603944,
          3.576185
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 46.32211499847472,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-uniform/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 24.1769,
        "first_ms": 3.508181,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.836246,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.526667,
          3.577382,
          3.441548,
          3.152703,
          3.618486
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 51.837353967130184,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 860.9278330113739,
        "first_ms": 326.3381200376898,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1331.903726910241,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          251.21875992044806,
          251.25836802180856,
          249.6653599664569,
          253.51341103669256,
          256.909942952916
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2606.8525910377502,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1058.3535520127043,
        "first_ms": 379.30849799886346,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1961.8152750190347,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          377.76770896743983,
          396.5117339976132,
          401.1584360850975,
          396.741695003584,
          385.3122319560498
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3444.3514940794557,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 861.8330849567428,
        "first_ms": 570.1573130208999,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1570.2100610360503,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          320.08097402285784,
          273.0738699901849,
          274.3758560391143,
          290.4946180060506,
          327.9139829101041
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3106.908045941964,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 25.259732,
        "first_ms": 4.308654,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.747876,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.172229,
          4.095154,
          4.11599,
          4.168486,
          5.169271
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37208,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 58.801071951165795,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 348.0595889268443,
        "first_ms": 709.7700369777158,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.78399098478258,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.742968991398811,
          5.555578973144293,
          5.9742790181189775,
          7.86728598177433,
          5.5080309975892305
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1338.1653559627011,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 154.99433397781104,
        "first_ms": 4.964747000485659,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.982605029828846,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.2307700710371137,
          2.278186031617224,
          3.358340007252991,
          2.9754230054095387,
          2.2811529925093055
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 236.31220299284905,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 676.9512370228767,
        "first_ms": 712.8181799780577,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 98.11214194633067,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.9349590186029673,
          3.594570909626782,
          4.0362890576943755,
          3.404259914532304,
          3.297767019830644
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1733.6387620307505,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.010453,
        "first_ms": 3.527336,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.531922,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.246803,
          3.788435,
          3.233169,
          9.976504,
          3.927476
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 57.582305977120996,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 48.277077,
        "first_ms": 5.615829,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 41.548932,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.219803,
          6.46389,
          9.787227,
          10.751074,
          8.297544
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36960,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 112.8646720899269,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 632.2460230439901,
        "first_ms": 559.76982798893,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 96.03140503168106,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.423299968242645,
          3.8739809533581138,
          2.158384071663022,
          2.823401940986514,
          3.37814399972558
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1537.9264579387382,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.859185,
        "first_ms": 2.827296,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.773971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.980064,
          3.209934,
          2.50392,
          5.844248,
          3.735216
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 49.205602961592376,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.487278,
        "first_ms": 2.390961,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.983,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.063099,
          2.010234,
          6.533361,
          2.071114,
          1.908172
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.975744978524745,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.041105,
        "first_ms": 2.328154,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.031983,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.106293,
          2.543009,
          2.267813,
          2.113017,
          3.724955
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.20401495695114,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 806.7208220018074,
        "first_ms": 350.50872492138296,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1213.6973770102486,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          228.15931495279074,
          230.8874810114503,
          229.27246091421694,
          230.57327000424266,
          229.14905496872962
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2455.8420029934496,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 800.7518569938838,
        "first_ms": 201.8714810255915,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1095.567067968659,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          203.30503396689892,
          207.135119009763,
          203.00246600527316,
          210.16937296371907,
          205.22145798895508
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2144.1912170266733,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 550.8236560272053,
        "first_ms": 2.5735750095918775,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 88.73267297167331,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.2496269084513187,
          3.6985200131312013,
          3.8525729905813932,
          2.0761480554938316,
          2.3625590838491917
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 735.5257770977914,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-skew/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 969.9543229071423,
        "first_ms": 620.8147339057177,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 131.55964098405093,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          11.277138954028487,
          8.79319105297327,
          6.735525093972683,
          5.208330927416682,
          6.848598946817219
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 2160.8161949552596,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 836.4941939944401,
        "first_ms": 221.33027494419366,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1182.877567014657,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          226.42951598390937,
          222.56641602143645,
          224.52221508137882,
          222.04406303353608,
          222.10642998106778
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2327.374243061058,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.893529,
        "first_ms": 2.948739,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.482714,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.377077,
          3.289989,
          2.74601,
          3.458294,
          3.24839
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 45.34057492855936,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/dense/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.582711,
        "first_ms": 3.556675,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.029589,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.995467,
          3.774871,
          4.022044,
          3.750032,
          3.467247
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36992,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 52.398249972611666,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 47.293862,
        "first_ms": 9.495025,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 29.320952,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.09537,
          7.723671,
          7.432659,
          3.032684,
          4.018089
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37080,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 93.86607201304287,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 552.385191898793,
        "first_ms": 5.268616019748151,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.34382100496441,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.6813920121639967,
          2.984542050398886,
          2.995487069711089,
          3.7618849892169237,
          2.5313899386674166
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 742.9061270086095,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-uniform/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.525786,
        "first_ms": 2.883147,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.645043,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.931945,
          2.063055,
          2.601784,
          2.853964,
          1.924868
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 38.85781590361148,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 26.124048,
        "first_ms": 3.903259,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.036223,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.559489,
          3.507681,
          3.170463,
          3.515221,
          3.713513
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 54.76689396891743,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.6005,
        "first_ms": 3.849976,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.398076,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.992808,
          3.966569,
          3.765358,
          3.75209,
          3.39602
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 47.78762301430106,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 40.936575,
        "first_ms": 4.276617,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.473236,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.931495,
          3.550906,
          3.693555,
          3.870479,
          3.893584
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 71.27843191847205,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 340.7541300402954,
        "first_ms": 694.3662739358842,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 29.222826939076185,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.383064970374107,
          5.679300986230373,
          4.588198033161461,
          5.438258056528866,
          3.5089170560240746
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1302.3972290102392,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.446315,
        "first_ms": 4.168224,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.839279,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.86755,
          3.373034,
          3.708676,
          3.969695,
          3.413966
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 47.938960953615606,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 24.093664,
        "first_ms": 4.37919,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.245507,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.196428,
          4.116798,
          4.10801,
          3.592828,
          4.201652
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37264,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 55.759264971129596,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.360516,
        "first_ms": 3.659364,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.725421,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.279982,
          2.671516,
          3.358118,
          3.425508,
          2.699999
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30228,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 46.843942953273654,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 682.7505419496447,
        "first_ms": 699.3943839333951,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 96.8311580363661,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.8016990292817354,
          3.0060959979891777,
          2.7580149471759796,
          2.9885219410061836,
          2.9064740519970655
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1766.7021059896797,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 29.427243,
        "first_ms": 7.443852,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 25.609,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.291409,
          5.321337,
          5.729275,
          3.999447,
          5.236306
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37544,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 70.64485701266676,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 27.259457,
        "first_ms": 4.290087,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.294599,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.913282,
          4.09071,
          6.37083,
          4.572592,
          6.306247
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36908,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 67.15894001536071,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 29.230764,
        "first_ms": 5.789594,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.743496,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.194531,
          3.944834,
          4.818006,
          3.601612,
          4.155426
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37076,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 64.37932001426816,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 1033.5015719756484,
        "first_ms": 1066.8266380671412,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2059.558439999819,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          418.8112650299445,
          393.3437200030312,
          397.30164397042245,
          380.1794769242406,
          384.2249009758234
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 4303.287063958123,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-skew/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 950.5750529933721,
        "first_ms": 270.19645692780614,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1418.357371003367,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          264.3661150941625,
          264.6482829004526,
          262.74010399356484,
          270.87121398653835,
          271.462781005539
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2715.2512189932168,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 26.475189,
        "first_ms": 5.034482,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 22.975283,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.351379,
          5.220778,
          3.338278,
          6.970951,
          4.073784
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37664,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 68.53769102599472,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-uniform/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 842.5609630066901,
        "first_ms": 220.5766710685566,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1175.449462956749,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          219.864921993576,
          223.75824104528874,
          220.95298604108393,
          221.92774701397866,
          221.10669291578233
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2294.1893379902467,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 637.643311987631,
        "first_ms": 2.7411580085754395,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 94.01784499641508,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.539088949561119,
          2.5764560559764504,
          2.5077570462599397,
          3.0382570112124085,
          3.8903510430827737
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 836.95167908445,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 813.9245599741116,
        "first_ms": 202.85464404150844,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1114.171089953743,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          206.36487298179418,
          201.75628294236958,
          207.15396897867322,
          225.30414501670748,
          205.89458500035107
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2199.6769360266626,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.357308,
        "first_ms": 2.189147,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.544207,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.878809,
          2.345023,
          5.650931,
          2.086944,
          2.283177
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.56708703562617,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.986252,
        "first_ms": 3.39524,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.643997,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.788916,
          4.041399,
          3.158089,
          4.410193,
          3.213739
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 46.79370508529246,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.806039,
        "first_ms": 4.133953,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.89194,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.524825,
          2.595851,
          4.10678,
          3.821591,
          2.351545
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 44.00600993540138,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1014.6536460379139,
        "first_ms": 561.6218680515885,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2359.7043440677226,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          467.67583896871656,
          488.23892208747566,
          466.5118099655956,
          465.2786289807409,
          467.89696300402284
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3998.0814940063283,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 856.5492699854076,
        "first_ms": 414.13608798757195,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1558.7782029761001,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          281.9625310366973,
          298.27956401277333,
          299.8073570197448,
          304.8772040056065,
          297.56491107400507
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2912.013405933976,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 298.11752995010465,
        "first_ms": 569.173575961031,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 25.91455401852727,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.5980320535600185,
          5.167854949831963,
          3.742421045899391,
          4.6626090770587325,
          3.7819549906998873
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1086.8656570091844,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 806.7864369368181,
        "first_ms": 205.5757089983672,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1112.0432489551604,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          205.62270190566778,
          205.92900994233787,
          214.92853097151965,
          212.4026739038527,
          205.9998350450769
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2172.180718043819,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 823.873660992831,
        "first_ms": 204.32733290363103,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1130.2939809393138,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          206.888917949982,
          206.38919796328992,
          205.22596908267587,
          219.28761305753142,
          222.72943297866732
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2206.2825600150973,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.520431,
        "first_ms": 3.690515,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.568042,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.192214,
          2.14037,
          2.170218,
          3.624852,
          2.122372
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 44.74819405004382,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 720.6852249801159,
        "first_ms": 700.1412380486727,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 107.96526400372386,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.872696961276233,
          3.043937962502241,
          7.798976032063365,
          3.1620230292901397,
          3.593459026888013
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1897.0726950792596,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 24.376794,
        "first_ms": 4.974468,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 22.421793,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.486114,
          3.761064,
          3.913614,
          4.264882,
          6.511587
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30068,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 60.438457992859185,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 25.562233,
        "first_ms": 3.873101,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.675928,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.498553,
          5.419867,
          5.830277,
          5.29737,
          6.003213
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 65.38574700243771,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 25.834635,
        "first_ms": 5.457373,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 30.57723,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.665003,
          7.643289,
          5.718051,
          5.247905,
          4.260777
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 69.59623494185507,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 30.369917,
        "first_ms": 7.422916,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.201074,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.842414,
          5.777823,
          5.882213,
          5.261405,
          5.399785
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37784,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 77.18731998465955,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 203.48867704160511,
        "first_ms": 4.684470943175256,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.935931040905416,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.465604946948588,
          3.2089450396597385,
          4.7441680217161775,
          5.699811968952417,
          4.785249009728432
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 318.6901309527457,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 35.916534,
        "first_ms": 4.8551,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.972174,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.808431,
          2.592792,
          3.170345,
          3.673881,
          6.181124
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 76.92085509188473,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 831.8919719895348,
        "first_ms": 536.43927606754,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 96.44404496066272,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.086204920895398,
          3.7997469771653414,
          2.707791980355978,
          3.739899955689907,
          3.665560041554272
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1750.5514639196917,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1016.5401879930869,
        "first_ms": 552.1659939549863,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1580.207281978801,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          403.37925602216274,
          401.6542120371014,
          255.8124699862674,
          262.5249660341069,
          253.38339898735285
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3235.6282010441646,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 997.561909025535,
        "first_ms": 424.7305759927258,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1149.844946921803,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          204.27906699478626,
          297.292627976276,
          199.7180419275537,
          211.53982495889068,
          182.01826501172036
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2707.7447490300983,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 44.714914,
        "first_ms": 1.990592,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 23.298676,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.701707,
          8.168071,
          3.198658,
          2.660206,
          3.262712
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30120,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 78.26253201346844,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 638.9409599360079,
        "first_ms": 3.919568029232323,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 96.29403206054121,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.41786106582731,
          3.4845350310206413,
          2.9222851153463125,
          5.5203179363161325,
          2.962085069157183
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 841.1678980337456,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.668815,
        "first_ms": 3.469792,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.526183,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.69577,
          3.582074,
          3.343713,
          4.043313,
          3.299297
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 45.14268005732447,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.67289,
        "first_ms": 2.937845,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.116965,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.287508,
          3.206373,
          2.747665,
          3.261684,
          3.32
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 44.781861011870205,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.144953,
        "first_ms": 3.051744,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.330496,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.272784,
          3.001279,
          2.634305,
          2.905972,
          3.196479
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30196,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 45.50786700565368,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1033.005633042194,
        "first_ms": 373.02993703633547,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1889.633646933362,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          382.475706981495,
          380.5600869236514,
          377.20433494541794,
          371.5086100855842,
          373.6494149779901
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3337.0917339343578,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/dense/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1035.1449369918555,
        "first_ms": 368.94300393760204,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1929.549525026232,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          374.47450403124094,
          386.7740340065211,
          390.82204597070813,
          388.79875699058175,
          384.1947439359501
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3375.566439004615,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 879.1770080570132,
        "first_ms": 326.76069100853056,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1403.4719750052318,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          265.9094480331987,
          259.2694979393855,
          260.6867420254275,
          255.14912803191692,
          291.0654579754919
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2694.0843280171975,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.61892,
        "first_ms": 3.613699,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.698129,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.903215,
          4.188515,
          2.872384,
          2.819348,
          4.574682
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30196,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 52.437462960369885,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 25.255626,
        "first_ms": 3.680213,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.534936,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.442513,
          3.757457,
          3.811251,
          4.88595,
          3.613411
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36920,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 57.561296969652176,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.267251,
        "first_ms": 4.192536,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.426061,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.880607,
          3.911282,
          3.803371,
          4.362115,
          3.868548
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 53.984081954695284,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 25.146657,
        "first_ms": 5.032746,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.326934,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.701329,
          4.340875,
          4.068039,
          3.673113,
          8.52006
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37080,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 62.08768906071782,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 712.462816038169,
        "first_ms": 675.0377389835194,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.41260197106749,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.793772961013019,
          2.881548018194735,
          2.9485179111361504,
          3.2288460060954094,
          3.2981010153889656
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1720.9215310867876,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/dense/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 158.60175096895546,
        "first_ms": 3.0232169665396214,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.14024001173675,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.040423918515444,
          2.342888037674129,
          2.960197045467794,
          2.0523270359262824,
          2.153867040760815
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 237.57737700361758,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 20.61854,
        "first_ms": 4.506666,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 20.052707,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.269141,
          4.192981,
          4.236205,
          4.316782,
          3.004103
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 51.092911045998335,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 34.570393,
        "first_ms": 2.319706,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.469076,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.636495,
          2.310464,
          2.87273,
          2.150213,
          2.16237
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 58.53338190354407,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 822.6592489518225,
        "first_ms": 430.2693100180477,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2061.5726159885526,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          755.2039310103282,
          354.70864397939295,
          358.1650099949911,
          234.20364293269813,
          277.2861300036311
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3410.466336994432,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 499.39931207336485,
        "first_ms": 302.4556420277804,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 685.3995519923046,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          118.7023869715631,
          119.90223696921021,
          116.8947669211775,
          118.31575201358646,
          126.33624405134469
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 1614.998042001389,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 625.8653270779178,
        "first_ms": 758.4389949915931,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 123.76482889521867,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          8.060817955993116,
          2.156337955966592,
          4.992709960788488,
          2.745043020695448,
          5.9860789915546775
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1900.531384977512,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1068.0750699248165,
        "first_ms": 316.9513629982248,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 936.7315279087052,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          186.60845106933266,
          186.1101109534502,
          186.30032090004534,
          186.10415502917022,
          187.2723480919376
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2436.9792139623314,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 567.4031910020858,
        "first_ms": 4.886450013145804,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 94.82391201891005,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.577291965484619,
          2.409695996902883,
          2.4885720340535045,
          3.66684899199754,
          3.8370470283553004
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 763.5478340089321,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-skew/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 559.8295980598778,
        "first_ms": 2.126873005181551,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 94.33795500081033,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.260059001855552,
          4.051134921610355,
          2.2581869270652533,
          2.242515911348164,
          2.3122639395296574
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 747.8677259059623,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.822204,
        "first_ms": 2.604556,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.094601,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.37489,
          3.806095,
          2.436026,
          2.133192,
          3.875604
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 43.06137398816645,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 30.199385,
        "first_ms": 5.435841,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.161638,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.410328,
          5.183579,
          7.956907,
          10.663085,
          5.718487
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 83.95881194155663,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 844.8149459436536,
        "first_ms": 317.10223597474396,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1191.837858990766,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          220.7046109251678,
          261.30698795896024,
          261.2489900784567,
          153.81875808816403,
          224.78248795960099
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2485.7867549872026,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 31.580044,
        "first_ms": 2.526337,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.298333,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.784688,
          1.726217,
          2.547034,
          6.046009,
          3.941084
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 57.497813017107546,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.087558,
        "first_ms": 4.078723,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.057722,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.034247,
          2.388089,
          4.596167,
          3.786141,
          4.230909
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37040,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 60.5643599992618,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 810.992719954811,
        "first_ms": 7.540380000136793,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 124.57151606213301,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          7.260441081598401,
          3.0214410508051515,
          6.506022997200489,
          7.173130987212062,
          9.302653023041785
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1116.313565056771,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 656.8041329737753,
        "first_ms": 307.604325003922,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1487.0382159715518,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          264.7936959983781,
          449.8585710534826,
          256.95237901527435,
          256.7182460334152,
          255.9082160005346
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2525.4572989651933,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 15.117473,
        "first_ms": 2.981153,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.53836,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.25664,
          2.224896,
          2.545497,
          2.494784,
          1.768919
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 34.85699894372374,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 966.9232500018552,
        "first_ms": 436.7812710115686,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1207.3766759131104,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          311.6178159834817,
          320.5612499732524,
          196.30709697958082,
          160.85315600503236,
          159.75373599212617
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2674.608047003858,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 16.24826,
        "first_ms": 3.067998,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 10.975208,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.701101,
          1.874217,
          3.074928,
          2.373158,
          1.69494
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 37.18145203310996,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 34.261549,
        "first_ms": 7.201457,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 30.477414,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          9.727015,
          7.787886,
          3.75245,
          4.241493,
          4.940021
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 77.40065502002835,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-uniform/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 34.696939,
        "first_ms": 4.125692,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 16.698861,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.161699,
          3.490188,
          3.279896,
          3.204079,
          3.025792
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 61.31716107483953,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 156.23073489405215,
        "first_ms": 4.639616003260016,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.361368962563574,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.196516005322337,
          2.4045760510489345,
          2.2215270437300205,
          3.6531949881464243,
          2.380346995778382
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 237.61772899888456,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.657082,
        "first_ms": 4.570444,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 57.358412,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.031006,
          13.724276,
          4.197771,
          16.662167,
          19.209879
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 90.26944194920361,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 923.4026060439646,
        "first_ms": 1028.8844930473715,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 87.02019392512739,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.144933937117457,
          5.216371035203338,
          3.162103006616235,
          2.142398036085069,
          3.245851956307888
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2312.970754923299,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-skew/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 25.668947,
        "first_ms": 5.003863,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 32.725449,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.627838,
          2.536796,
          7.819731,
          9.253404,
          7.089316
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 74.485121993348,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 39.27173,
        "first_ms": 13.284433,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 10.440099,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.451803,
          1.805588,
          1.691754,
          2.456176,
          1.793436
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 72.90958997327834,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 580.4265760816634,
        "first_ms": 259.57508594729006,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1046.73254606314,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          250.3103750059381,
          206.8259980296716,
          174.71959895920008,
          201.69940602499992,
          151.65546396747231
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 1973.4031510306522,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1048.7435009563342,
        "first_ms": 300.57312606368214,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1461.1814219970256,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          245.22225908003747,
          297.8758519748226,
          255.12935500591993,
          298.305316013284,
          361.8746669963002
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 2864.0232658945024,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 29.194561,
        "first_ms": 5.397751,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 37.870544,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.546763,
          6.376551,
          9.87916,
          6.178162,
          9.425289
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 82.65505696181208,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 28.863038,
        "first_ms": 6.747362,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.984878,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.63762,
          6.272844,
          3.942813,
          4.110097,
          1.795604
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 59.685845975764096,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 1000.2099169651046,
        "first_ms": 1139.735987992026,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1986.9414790300652,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          834.9793370580301,
          275.2675209194422,
          271.52844693046063,
          266.59578492399305,
          256.5505140228197
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 4203.982756938785,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 544.6244720369577,
        "first_ms": 527.0692380145192,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 88.09592993929982,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.9018489187583327,
          3.295646049082279,
          2.6222749147564173,
          2.061723032966256,
          2.1426730090752244
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1371.5532220667228,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 532.7363610267639,
        "first_ms": 3.4598540514707565,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.65235790330917,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.792992047034204,
          3.333591972477734,
          2.156652044504881,
          2.734281006269157,
          3.1160860089585185
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 725.8310039760545,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.350848,
        "first_ms": 3.84972,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.097405,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.049178,
          3.576724,
          3.827015,
          3.076238,
          3.550007
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36992,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 50.167865003459156,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/dense/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 40.637418,
        "first_ms": 7.196609,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 43.549546,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.324789,
          10.469848,
          6.481545,
          6.503765,
          12.748951
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36964,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 105.69372994359583,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 23.67041,
        "first_ms": 2.722524,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.551337,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.410358,
          2.388214,
          3.62873,
          2.283575,
          2.340752
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 47.81755502335727,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 799.2171699879691,
        "first_ms": 206.45778393372893,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1084.0430189855397,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          204.95924702845514,
          204.87920194864273,
          204.92025394923985,
          200.64173196442425,
          203.16889008972794
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2139.5387169905007,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.053863,
        "first_ms": 3.343119,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.570831,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.857483,
          3.610664,
          3.767704,
          4.77911,
          4.535045
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37048,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 53.185936994850636,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 19.762177,
        "first_ms": 3.3145,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 48.120603,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.110667,
          14.964834,
          9.042544,
          10.965645,
          5.993022
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 76.71774399932474,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 26.164874,
        "first_ms": 9.345071,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 30.76248,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.396762,
          7.261243,
          7.149436,
          6.221011,
          2.274835
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 75.36352693568915,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 36.373508,
        "first_ms": 5.95935,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.223701,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.170411,
          4.885344,
          5.452821,
          5.278547,
          6.178373
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30176,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 76.14026893861592,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 27.199665,
        "first_ms": 3.731325,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 27.432287,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.871124,
          5.326149,
          6.985722,
          8.003707,
          3.212008
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 66.28097395878285,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 808.0680640414357,
        "first_ms": 220.10205092374235,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 979.375306982547,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          221.0306168999523,
          165.30814999714494,
          222.36070199869573,
          158.1603439990431,
          150.44158906675875
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2078.232315951027,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 744.6038029156625,
        "first_ms": 411.34415194392204,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1319.1736959852278,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          217.10556210018694,
          202.07455498166382,
          205.3500449983403,
          263.8263739645481,
          370.43733603786677
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2603.4939870005473,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 35.744056,
        "first_ms": 6.376679,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.994156,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.071634,
          3.890869,
          1.905671,
          4.988088,
          3.663354
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 68.99314594920725,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 846.6367489891127,
        "first_ms": 281.2260090140626,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 903.4463020507246,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          158.52572408039123,
          173.03442698903382,
          172.4721910431981,
          165.7213579164818,
          170.97165901213884
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2131.766808917746,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 430.58611801825464,
        "first_ms": 3.6685270024463534,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 59.507301077246666,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          1.981709967367351,
          1.844611018896103,
          2.66732694581151,
          2.855060971342027,
          1.9066919339820743
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 587.0800920529291,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 48.421309,
        "first_ms": 5.063247,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.731122,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.741913,
          1.614039,
          5.927178,
          1.717386,
          6.506933
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 88.46253098454326,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/dense/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 14.22252,
        "first_ms": 3.041804,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 16.964997,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.191364,
          3.74984,
          3.04146,
          3.387059,
          3.568083
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 38.23802201077342,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 709.9396679550409,
        "first_ms": 518.0587419308722,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1947.2922480199486,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          336.50987793225795,
          371.22492596972734,
          431.9795210612938,
          407.6833510771394,
          396.87857404351234
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3233.285915106535,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 1014.5716250408441,
        "first_ms": 229.71244505606592,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1244.486914947629,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          240.06267404183745,
          219.77153897751123,
          237.2390910750255,
          238.6613111011684,
          239.9512919364497
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2568.7100359937176,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 990.4921599663794,
        "first_ms": 365.0206249440089,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1232.1029000449926,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          217.80251897871494,
          239.03143603820354,
          227.42323798593134,
          270.5367690650746,
          274.5585530065
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2689.5686300704256,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 286.61245200783014,
        "first_ms": 537.6138689462095,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.620185955427587,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.6174439592286944,
          4.925298970192671,
          3.589969011954963,
          4.224776057526469,
          3.5439470084384084
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1051.2462359620258,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-uniform/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.387532,
        "first_ms": 3.729035,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.172655,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.918219,
          1.911572,
          1.938449,
          4.067532,
          2.066343
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.61139200348407,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1061.6817019181326,
        "first_ms": 353.3367660129443,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1762.1420549694449,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          352.3228069534525,
          351.50334099307656,
          351.362303015776,
          351.00913094356656,
          351.5299829887226
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3211.947741918266,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.740411,
        "first_ms": 3.014204,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.445302,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.329917,
          2.342108,
          1.959969,
          2.002282,
          3.498224
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.5922329518944,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.07867,
        "first_ms": 3.144265,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.681884,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.293832,
          2.345968,
          2.909465,
          3.297913,
          2.348851
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 40.36408301908523,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 557.6248299330473,
        "first_ms": 4.477145033888519,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 92.03288296703249,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.8540909299626946,
          2.8826750349253416,
          2.3172979708760977,
          2.4316110648214817,
          2.3209049832075834
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 743.0426859064028,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-skew/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 592.4599739955738,
        "first_ms": 356.123381992802,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 97.71096392069012,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.3029709700495005,
          2.833493985235691,
          2.205890021286905,
          2.3444080725312233,
          3.6616820143535733
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1253.1604469986632,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 137.44584808591753,
        "first_ms": 3.766078967601061,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.963850987143815,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.558872966095805,
          2.0655780099332333,
          2.149599022231996,
          3.139890031889081,
          2.104114042595029
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 209.00903001893312,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.128703,
        "first_ms": 2.907487,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.345635,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.106188,
          2.139136,
          2.954457,
          2.694109,
          2.144836
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.359777075238526,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.604195,
        "first_ms": 3.091262,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.338574,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.145514,
          2.329285,
          2.447377,
          2.982129,
          2.135073
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.04040200076997,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-skew/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.563464,
        "first_ms": 5.259489,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.215852,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.288622,
          2.342319,
          2.20141,
          3.808975,
          2.267839
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 46.14437907002866,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.2987,
        "first_ms": 3.657023,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.791588,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.181479,
          3.902926,
          3.636977,
          3.180373,
          3.869056
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36968,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 53.688714979216456,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.233293,
        "first_ms": 2.834481,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.638272,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.901117,
          2.218485,
          2.007956,
          2.093509,
          2.941309
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.044811067171395,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/dense/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 1,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 289.4334470620379,
        "first_ms": 347.503372002393,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.789524920284748,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.140379023738205,
          6.550149992108345,
          4.581838031299412,
          4.237179062329233,
          4.37041197437793
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 884.2414440587163,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 551.0452740127221,
        "first_ms": 2.1797389490529895,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.47284198459238,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.080510021187365,
          2.091175061650574,
          3.5157629754394293,
          2.0224059699103236,
          2.0011940505355597
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 733.3074229536578,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.836543,
        "first_ms": 3.357809,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.630459,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.986453,
          3.68217,
          3.933953,
          3.773776,
          3.220568
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 46.079184976406395,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.360563,
        "first_ms": 1.969455,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 10.981385,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.484031,
          2.742458,
          1.842409,
          1.79844,
          1.839805
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.484026026912034,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 838.6926769744605,
        "first_ms": 303.2615160336718,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1261.7459419416264,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          238.18647197913378,
          236.73937795683742,
          239.80041197501123,
          239.33644709177315,
          238.91398205887526
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2470.461779041216,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.383064,
        "first_ms": 2.384666,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.822513,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.002325,
          1.937172,
          3.594265,
          2.005666,
          1.998396
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 41.25466605182737,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 597.8061739588156,
        "first_ms": 556.7258630180731,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 92.05153398215771,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.1190510597079992,
          2.8861960163339972,
          2.113310038112104,
          2.168947015888989,
          2.88838311098516
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1452.0372370025143,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 833.8831979781389,
        "first_ms": 389.7057439899072,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 765.9105970524251,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          149.91347503382713,
          140.7481109490618,
          128.84709099307656,
          132.399957976304,
          126.84042495675385
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2111.9485599920154,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 36.637294,
        "first_ms": 6.913029,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.712394,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.621225,
          10.090424,
          7.184503,
          6.193362,
          6.602662
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36960,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 94.57203594502062,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 31.174379,
        "first_ms": 8.62937,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 37.513977,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.476453,
          7.557404,
          7.899706,
          8.175983,
          7.360214
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 89.94366403203458,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 802.2842149948701,
        "first_ms": 320.8531039999798,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1187.6707749906927,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          204.41467594355345,
          208.8597909314558,
          206.71385910827667,
          205.8118119603023,
          299.0299239754677
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2431.8762209732085,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 36.842124,
        "first_ms": 8.918889,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 37.950245,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          8.920921,
          6.859333,
          4.956617,
          6.863624,
          9.953164
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 92.6249319454655,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 34.563935,
        "first_ms": 4.188525,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.497408,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.071883,
          3.462953,
          7.952576,
          2.514784,
          6.46237
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37024,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 71.82724704034626,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-uniform/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.917195,
        "first_ms": 3.372172,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.804241,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.532735,
          3.244385,
          2.83419,
          3.161407,
          2.644168
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 47.15418501291424,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1043.1186240166426,
        "first_ms": 728.5814250353724,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1844.4454930722713,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          442.97629594802856,
          326.8278749892488,
          349.7109250165522,
          350.69663904141635,
          369.795780046843
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3705.052323988639,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.753406,
        "first_ms": 4.144748,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.093961,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.968625,
          3.897089,
          3.273724,
          2.784346,
          7.819243
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 56.103788083419204,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 30.395792,
        "first_ms": 13.812983,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 39.977181,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.999458,
          7.091351,
          6.299131,
          10.565039,
          8.989532
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 92.1515979571268,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 36.487292,
        "first_ms": 5.19335,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.711464,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.95494,
          2.639311,
          6.051387,
          6.152548,
          2.656
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 72.79161701444536,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 28.450371,
        "first_ms": 3.971034,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.851975,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.107835,
          3.404414,
          5.026683,
          3.870254,
          4.115651
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 61.77849601954222,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 162.42878697812557,
        "first_ms": 5.498858983628452,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 22.79154199641198,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.5527489380910993,
          3.916808986105025,
          3.408852033317089,
          3.296158043667674,
          3.873956040479243
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 254.5696940505877,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.881725,
        "first_ms": 3.638612,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.113833,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.536076,
          3.365843,
          4.197602,
          4.358718,
          3.633417
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36940,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 63.13213403336704,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.526585,
        "first_ms": 3.961086,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.562,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.66258,
          2.526443,
          3.068905,
          3.277365,
          2.683204
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30072,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 49.18149102013558,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.123674,
        "first_ms": 3.248133,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.437171,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.126408,
          3.983418,
          2.683291,
          2.604453,
          3.650209
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30248,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 47.64450108632445,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-skew/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.090333,
        "first_ms": 2.786158,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.374644,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.069697,
          2.614178,
          3.014278,
          3.618675,
          2.647274
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 44.53786194790155,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 940.5562109313905,
        "first_ms": 398.3847090275958,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1243.989193928428,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          257.8932299511507,
          218.2613230543211,
          232.2002020664513,
          232.32888791244477,
          235.07167398929596
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2681.1874179402366,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 828.4659079508856,
        "first_ms": 220.94305499922484,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1141.6038449387997,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          213.14110490493476,
          214.38957599457353,
          215.70562198758125,
          215.07871500216424,
          218.5412289109081
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2248.1682780198753,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 668.8254770124331,
        "first_ms": 654.0229379897937,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.77671996783465,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.8810410294681787,
          2.643786952830851,
          3.092742059379816,
          3.162472043186426,
          2.8169670840725303
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1660.633870982565,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 718.2907300302759,
        "first_ms": 693.4471989516169,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 91.23434196226299,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.02252103574574,
          2.717479015700519,
          2.8200630331411958,
          2.8166320407763124,
          2.703494974412024
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1742.6171089755371,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.249623,
        "first_ms": 2.922732,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.806851,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.77335,
          3.752479,
          2.607619,
          2.663505,
          3.669182
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 45.94771901611239,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1021.1963800247759,
        "first_ms": 367.06798803061247,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2046.8280469067395,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          363.9866429148242,
          545.3514020191506,
          361.9829440722242,
          383.1118729431182,
          387.8698870539665
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3480.0550549989566,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.541297,
        "first_ms": 3.919035,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.987127,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.972315,
          3.9035,
          3.511156,
          4.401796,
          3.616247
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 49.52745989430696,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.947815,
        "first_ms": 3.69022,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.659451,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.73075,
          3.012398,
          3.422338,
          2.585586,
          2.595633
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 48.2963080285117,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1056.3077599508688,
        "first_ms": 664.2848569899797,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2710.059560951777,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          541.1484059877694,
          540.122331934981,
          542.7061399677768,
          547.1321659861133,
          534.54993898049
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 4506.156389019452,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/dense/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.976909,
        "first_ms": 4.196542,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.539958,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.090096,
          4.173696,
          4.278599,
          4.505999,
          3.467056
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36956,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 56.66372203268111,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 653.2785840099677,
        "first_ms": 4.3180210050195456,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 115.06133899092674,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.3644699724391103,
          3.3956520492210984,
          4.738472984172404,
          4.688328015618026,
          4.608931019902229
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 951.92365499679,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 35.815807,
        "first_ms": 8.479811,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 35.872051,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.594893,
          6.9625,
          5.870027,
          10.856534,
          6.038294
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 85.10598994325846,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 46.318174,
        "first_ms": 9.989845,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.045191,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.397446,
          5.672615,
          6.880208,
          3.065065,
          3.598051
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 90.59314208570868,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.735886,
        "first_ms": 4.457235,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.545542,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.633756,
          3.922909,
          4.558171,
          4.00286,
          4.392965
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 53.84661303833127,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1053.193124011159,
        "first_ms": 371.98938406072557,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2282.467441051267,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          468.55706500355154,
          453.0451629543677,
          452.10781309287995,
          444.12563496734947,
          460.03852889407426
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3776.0797599330544,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 1014.5496610784903,
        "first_ms": 283.1257360521704,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1430.00443209894,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          274.2250619921833,
          272.47065398842096,
          272.7661509998143,
          270.682470058091,
          268.8993139890954
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2804.0589810116217,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 858.3934229100123,
        "first_ms": 985.7396259903908,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 135.89963293634355,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          11.09870697837323,
          3.6425490397959948,
          6.634042016230524,
          9.525290923193097,
          9.823066066019237
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2428.804542985745,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 41.931776,
        "first_ms": 10.951834,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.289059,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.627879,
          3.066464,
          9.8033,
          5.779606,
          2.737786
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30248,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 96.071726991795,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.292873,
        "first_ms": 5.334211,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.125786,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.233991,
          4.962804,
          3.431529,
          3.767062,
          3.709961
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37072,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 54.4741420308128,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.09599,
        "first_ms": 2.880559,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.260444,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.396226,
          3.959799,
          2.358049,
          2.327713,
          3.745207
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 41.437151026912034,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.026649,
        "first_ms": 3.518399,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.925789,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.013681,
          2.330534,
          3.390133,
          3.16854,
          2.553094
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 41.84831399470568,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 793.9256140962243,
        "first_ms": 357.43021802045405,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1357.9977990593761,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          249.90877893287688,
          256.70882500708103,
          250.3277900395915,
          261.1036760499701,
          240.07835902739316
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2617.1919479966164,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 928.6058379802853,
        "first_ms": 423.1955960858613,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2193.656934075989,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          367.72910598665476,
          363.20489400532097,
          460.5883789481595,
          497.58859793655574,
          500.1524150138721
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3647.204078035429,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 567.731534014456,
        "first_ms": 317.1247439458966,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 86.06860402505845,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.0526479929685593,
          1.880978001281619,
          2.724541933275759,
          1.923121977597475,
          2.7116460260003805
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1181.5596689702943,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 807.4737410061061,
        "first_ms": 456.8289259914309,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1511.8965110741556,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          271.6808350523934,
          278.8472840329632,
          278.56240095570683,
          279.1027220664546,
          333.9501660084352
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2855.31026404351,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.687765,
        "first_ms": 2.933691,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.744416,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.624792,
          3.061878,
          2.426739,
          3.118124,
          2.162992
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 50.80294399522245,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.182418,
        "first_ms": 3.638127,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.658296,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.892593,
          3.254643,
          3.151586,
          5.13698,
          3.202592
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37144,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 54.424536996521056,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.470517,
        "first_ms": 2.287938,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.363596,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.938269,
          2.741224,
          2.267521,
          2.202057,
          1.919407
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.90004200022668,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 552.9181539313868,
        "first_ms": 3.6290899151936173,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.77855097409338,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.506497945636511,
          2.2556909825652838,
          3.1200250377878547,
          3.702835994772613,
          2.2317530820146203
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 742.8843819070607,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.784716,
        "first_ms": 3.911714,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.50453,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.859281,
          3.088101,
          3.723961,
          3.75434,
          3.058796
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37656,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 52.18559899367392,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.373428,
        "first_ms": 2.717312,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.367624,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.010935,
          2.015555,
          1.971457,
          2.233666,
          3.815641
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.933287913911045,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.009136,
        "first_ms": 2.411806,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.717069,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.074886,
          2.828844,
          3.108502,
          2.306209,
          2.119898
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30168,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 41.05019196867943,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.91627,
        "first_ms": 5.236543,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.385334,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.112608,
          4.111141,
          3.940563,
          3.114429,
          4.085106
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 54.42363198380917,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 804.2345619760454,
        "first_ms": 380.0397770246491,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1320.3349250834435,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          203.95450096111745,
          245.91160891577601,
          255.06723404396325,
          278.96147803403437,
          249.02143294457346
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2603.8185940124094,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.958229,
        "first_ms": 4.372213,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.182926,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.097071,
          2.125028,
          2.196098,
          4.039126,
          2.442144
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.73081703670323,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.354524,
        "first_ms": 3.671325,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.498743,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.010446,
          3.673881,
          3.212502,
          4.069478,
          3.498168
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37772,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 54.96083002071828,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-skew/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 812.6785058993846,
        "first_ms": 2.3798459442332387,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.65613499842584,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.2353510139510036,
          3.0210400000214577,
          2.223026007413864,
          2.948397072032094,
          2.192219952121377
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1032.5747489696369,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.755976,
        "first_ms": 2.91337,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.633262,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.574473,
          4.790262,
          2.838453,
          3.570033,
          3.337001
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 45.09053402580321,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.686001,
        "first_ms": 2.23011,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.665236,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.784907,
          2.92613,
          2.051348,
          2.016168,
          2.584867
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.477545000612736,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1029.708594083786,
        "first_ms": 451.9929299131036,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1826.990419998765,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          346.1481010308489,
          376.2461310252547,
          397.1546719549224,
          374.0532100200653,
          329.15910496376455
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3405.059692915529,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.820881,
        "first_ms": 2.863855,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 16.312827,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.229238,
          3.298884,
          2.494308,
          3.312449,
          3.254738
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 43.12821105122566,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-uniform/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 774.1613830439746,
        "first_ms": 198.71657597832382,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1073.2990000396967,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          202.33568304684013,
          201.54977403581142,
          200.96849696710706,
          200.41312999092042,
          202.45201897341758
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2092.782710911706,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 996.2836620397866,
        "first_ms": 331.3075009500608,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1730.2183239953592,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          335.71500796824694,
          347.89381292648613,
          355.0407870206982,
          348.470023018308,
          338.67548406124115
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3093.079084996134,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.041854,
        "first_ms": 2.116237,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.722229,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.579786,
          2.096527,
          2.123051,
          2.085418,
          3.539513
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 43.11455797869712,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.906504,
        "first_ms": 16.104029,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 33.248009,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.19573,
          2.423544,
          12.080522,
          8.88841,
          5.878234
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 74.21900099143386,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 37.1812,
        "first_ms": 8.365653,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 38.226018,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          10.730877,
          7.352912,
          5.853954,
          7.085494,
          6.723165
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 95.44078598264605,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-uniform/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 39.130042,
        "first_ms": 5.282273,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 23.363934,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.235138,
          1.88797,
          5.132324,
          5.600513,
          5.057274
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 76.76216203253716,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 47.672745,
        "first_ms": 9.24204,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 51.907718,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.840786,
          11.89189,
          9.866352,
          11.909592,
          10.889149
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 116.42312107142061,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 284.4082290539518,
        "first_ms": 556.0080730356276,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.549157919362187,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.611797980032861,
          4.1585080325603485,
          4.399273078888655,
          4.0612779557704926,
          4.497573943808675
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1085.931709036231,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.230996,
        "first_ms": 2.229801,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.984751,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.979997,
          2.725679,
          2.123909,
          2.01321,
          2.872967
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.84672394674271,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-skew/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 805.0500630633906,
        "first_ms": 199.19938896782696,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1080.0216839415953,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          200.25739108677953,
          201.92403497640043,
          205.7046270929277,
          201.29502704367042,
          203.6477830260992
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2131.54441898223,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 19.08221,
        "first_ms": 4.922275,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.59959,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.299905,
          4.651411,
          3.25293,
          3.325024,
          5.036162
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 50.191963906399906,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 548.8918039482087,
        "first_ms": 4.3414050014689565,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.44687200710177,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.869693959131837,
          2.1241329377517104,
          2.150489017367363,
          2.9385959496721625,
          3.0475700041279197
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 734.9753350717947,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-skew/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.398503,
        "first_ms": 3.503782,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.436253,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.37065,
          2.97891,
          2.365518,
          3.250054,
          2.975564
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 40.31744203530252,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/dense/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1232.2013749508187,
        "first_ms": 438.08914406690747,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1998.68494796101,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          433.4431579336524,
          434.3019869411364,
          433.5866760229692,
          248.50433401297778,
          437.03902303241193
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 3750.9877840057015,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 564.6512450184673,
        "first_ms": 543.7732299324125,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.00687091611326,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.1699109822511673,
          2.154591027647257,
          2.606026013381779,
          2.797149005346,
          3.796944976784289
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1445.7830809988081,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.174251,
        "first_ms": 4.333341,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.787974,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.431078,
          5.004761,
          3.094256,
          4.706497,
          2.533278
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37272,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 51.2906500371173,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 542.8147590719163,
        "first_ms": 3.6335319746285677,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.06143100466579,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.8229840099811554,
          3.2167279860004783,
          3.2066229032352567,
          2.3493269691243768,
          2.4165509967133403
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 727.3729429580271,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 958.6752449395135,
        "first_ms": 382.30403198394924,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1169.7109499946237,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          157.71950501948595,
          203.01256200764328,
          306.44711304921657,
          286.9004450039938,
          158.8985079433769
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2566.7424430139363,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/dense/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 288.7423620559275,
        "first_ms": 256.68638793285936,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 30.193922924809158,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.099246066994965,
          3.7926159566268325,
          2.8674990171566606,
          6.295663071796298,
          11.198604013770819
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 998.7101439619437,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 872.83041595947,
        "first_ms": 358.75636897981167,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1258.2618159940466,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          253.6885350709781,
          234.77956501301378,
          265.92622604221106,
          261.4891950506717,
          239.6793900988996
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2600.457286927849,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 816.3534270133823,
        "first_ms": 146.09112800098956,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 857.9865980427712,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          227.60133200790733,
          146.64700406137854,
          148.30703404732049,
          146.57200395595282,
          146.09801000915468
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1875.4797199508175,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 502.7309429133311,
        "first_ms": 312.66412895638496,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1313.8963069068268,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          222.60237100999802,
          209.3950529815629,
          209.08249798230827,
          201.1630500201136,
          414.9037019815296
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2225.4811740713194,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 33.180417,
        "first_ms": 5.653449,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 34.152632,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          9.704766,
          5.904779,
          5.933008,
          5.885704,
          6.241188
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 78.3908748999238,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 684.6210670191795,
        "first_ms": 372.779848985374,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 988.6671709828079,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          192.40787404123694,
          192.3178219003603,
          195.77698793727905,
          203.10004393104464,
          201.2430370086804
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2177.689258940518,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 867.5224370090291,
        "first_ms": 219.64827994816005,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1167.862731963396,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          210.89521690737456,
          227.9725450789556,
          220.91466502752155,
          223.84468698874116,
          214.5989560522139
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2317.8527909331024,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 315.6536449678242,
        "first_ms": 603.4000850049779,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.55634400434792,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.7688789889216423,
          5.241650971584022,
          3.669162979349494,
          5.441225948743522,
          3.63774795550853
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1161.474140943028,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 42.468776,
        "first_ms": 8.332208,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 31.693612,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.824806,
          10.883932,
          3.298283,
          5.313535,
          7.353008
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37056,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 96.30613098852336,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 1016.7022849200293,
        "first_ms": 239.4576600054279,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 833.3146509248763,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          228.75271702650934,
          139.06475494150072,
          142.54711801186204,
          140.30441700015217,
          142.14697398710996
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2149.062279961072,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/dense/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 167.09271899890155,
        "first_ms": 2.344088046811521,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.31662600953132,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          1.785650965757668,
          2.6637460105121136,
          1.6987570561468601,
          2.496712957508862,
          2.60363204870373
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 240.99740595556796,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 814.2680620076135,
        "first_ms": 311.4938939688727,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 916.6005250299349,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          163.49399695172906,
          161.91788704600185,
          179.6064879745245,
          170.79535697121173,
          163.67381892632693
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2159.7177360672504,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.740355,
        "first_ms": 5.152524,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 23.47404,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.544164,
          5.311728,
          4.907231,
          4.70269,
          3.974582
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 59.40498295240104,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 170.2088430756703,
        "first_ms": 6.801848066970706,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 26.625751983374357,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.6110330149531364,
          4.914568038657308,
          3.8425150560215116,
          3.82275995798409,
          4.893422010354698
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 267.5405410118401,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 334.76529899053276,
        "first_ms": 596.2700039381161,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 30.131667968817055,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.357360001653433,
          4.674828029237688,
          5.825061001814902,
          4.533106926828623,
          4.799041897058487
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1210.412151995115,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.754677,
        "first_ms": 3.034436,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.941695,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.042774,
          2.670913,
          2.061518,
          2.858838,
          2.030335
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 40.2017809683457,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.335357,
        "first_ms": 4.27406,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.650435,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.93908,
          4.323739,
          3.12429,
          2.970427,
          4.271625
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36928,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 49.20518305152655,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 123.56113793794066,
        "first_ms": 3.201080020517111,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.674323986284435,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.9927389696240425,
          3.2631580252200365,
          1.9947139080613852,
          1.910642022266984,
          3.089384059421718
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 190.49501908011734,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.283317,
        "first_ms": 2.123412,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.690656,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.965558,
          1.929551,
          3.425125,
          1.956236,
          2.159312
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.38308509532362,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.339041,
        "first_ms": 3.226498,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.32151,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.809075,
          1.979473,
          1.924821,
          2.728651,
          2.600132
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30076,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 39.84008799307048,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.441243,
        "first_ms": 2.323237,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.595488,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.842509,
          2.102107,
          1.978604,
          1.957106,
          3.456619
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30232,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 42.10106306709349,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 522.636073990725,
        "first_ms": 2.4747319985181093,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 86.14073996432126,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.384372055530548,
          2.9078529914841056,
          2.1628899266943336,
          2.215675078332424,
          2.666320069693029
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 696.8742249300703,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.908935,
        "first_ms": 3.117233,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.426142,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.253045,
          2.979439,
          3.706023,
          3.411235,
          3.058455
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36936,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 50.319769070483744,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.546421,
        "first_ms": 3.089333,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.183978,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.287265,
          2.290632,
          2.34176,
          3.43232,
          2.368479
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 39.65973109006882,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.450853,
        "first_ms": 3.022215,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.290442,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.222507,
          2.970932,
          2.190545,
          2.81718,
          2.455016
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 38.56546606402844,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 797.2811000654474,
        "first_ms": 350.7198339793831,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1833.0392059870064,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          310.3911690413952,
          292.1418610494584,
          260.88584505487233,
          337.72533701267093,
          556.4206739654765
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3088.6318699922413,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/dense/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 17.803371,
        "first_ms": 3.091486,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.193625,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.333511,
          3.666863,
          3.398555,
          3.717023,
          3.046008
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 43.304119957610965,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 521.1741799721494,
        "first_ms": 3.362608957104385,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 87.71893300581723,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.6675359588116407,
          3.3548660576343536,
          2.232267986983061,
          2.2814590483903885,
          3.0325689585879445
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 702.8694779146463,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 559.5933699514717,
        "first_ms": 531.9862050237134,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 86.09476906713098,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.0773709984496236,
          2.237839042209089,
          3.5461559891700745,
          2.466991078108549,
          2.010046038776636
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1374.7666700510308,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-skew/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 544.8807720094919,
        "first_ms": 316.5162520017475,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 87.61020307429135,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.8704769210889935,
          2.0669150399044156,
          2.4629109539091587,
          2.513158950023353,
          2.7575409039855003
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1162.0835840003565,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 781.6206680145115,
        "first_ms": 369.1305819666013,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1450.5763299530372,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          337.78136607725173,
          243.286824086681,
          262.56556098815054,
          254.4249630300328,
          267.26148603484035
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2688.004603027366,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "timing",
      "launch": 2,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 792.347960988991,
        "first_ms": 202.45332701597363,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1060.1294020889327,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          204.8534289933741,
          197.86071602720767,
          196.0096909897402,
          199.28774796426296,
          198.19804502185434
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2103.3122439403087,
      "sampled_peak_tree_pss_bytes": null,
      "max_observed_processes": null,
      "pss_samples": null,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1022.5996400695294,
        "first_ms": 342.5791460322216,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 672.4357600323856,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          337.7038659527898,
          333.0485710175708
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 2183.1005099229515,
      "sampled_peak_tree_pss_bytes": 45383680,
      "max_observed_processes": 1,
      "pss_samples": 184,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 19.431353,
        "first_ms": 3.317429,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 7.83402,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.566457,
          3.253947
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 149.00557196233422,
      "sampled_peak_tree_pss_bytes": 34269184,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 833.7714459048584,
        "first_ms": 490.9758960129693,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 794.7164910146967,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          466.7156810173765,
          292.10504703223705
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2355.5956119671464,
      "sampled_peak_tree_pss_bytes": 245575680,
      "max_observed_processes": 6,
      "pss_samples": 158,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1242.3733239993453,
        "first_ms": 405.31142603140324,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 807.7889840351418,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          398.46923598088324,
          408.1469669472426
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 2618.757591000758,
      "sampled_peak_tree_pss_bytes": 45446144,
      "max_observed_processes": 1,
      "pss_samples": 212,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 49.565784,
        "first_ms": 7.520022,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 20.54954,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          12.436746,
          8.097761
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36964,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 197.6454040268436,
      "sampled_peak_tree_pss_bytes": 34436096,
      "max_observed_processes": 1,
      "pss_samples": 17,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 847.212471999228,
        "first_ms": 486.16918502375484,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 633.4823879878968,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          275.2055450109765,
          323.98646394722164
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2192.668357049115,
      "sampled_peak_tree_pss_bytes": 179389440,
      "max_observed_processes": 4,
      "pss_samples": 163,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 21.055813,
        "first_ms": 2.670504,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 6.655661,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.414505,
          4.042228
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 145.53049893584102,
      "sampled_peak_tree_pss_bytes": 26609664,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 893.1933660060167,
        "first_ms": 736.4121619611979,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.046345031820238,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          1.9029730465263128,
          1.7206399934366345
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2094.935402041301,
      "sampled_peak_tree_pss_bytes": 173091840,
      "max_observed_processes": 1,
      "pss_samples": 161,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.849708,
        "first_ms": 4.700348,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 7.256534,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.289275,
          3.761173
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 146.21485595125705,
      "sampled_peak_tree_pss_bytes": 26580992,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-skew/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 24.470993,
        "first_ms": 2.475868,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 6.596477,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.957652,
          2.447602
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 147.0023230649531,
      "sampled_peak_tree_pss_bytes": 26568704,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 536.4486129255965,
        "first_ms": 3.978213993832469,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 34.96205701958388,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.4015820818021894,
          2.9512899927794933
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 777.1881079534069,
      "sampled_peak_tree_pss_bytes": 83603456,
      "max_observed_processes": 1,
      "pss_samples": 66,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 22.347752,
        "first_ms": 3.883969,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.637712,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.944773,
          3.685574
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37204,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 148.26683094725013,
      "sampled_peak_tree_pss_bytes": 34505728,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 21.709073,
        "first_ms": 3.192309,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.690269,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.957125,
          4.725839
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37524,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 145.62249404843897,
      "sampled_peak_tree_pss_bytes": 34907136,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.567385,
        "first_ms": 3.237876,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.239822,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.071824,
          2.065195
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 142.05599401611835,
      "sampled_peak_tree_pss_bytes": 26898432,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 40.932692,
        "first_ms": 7.413117,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.074517,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          11.210796,
          6.752824
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 181.20854208245873,
      "sampled_peak_tree_pss_bytes": 26788864,
      "max_observed_processes": 1,
      "pss_samples": 16,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/dense/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1089.6977670490742,
        "first_ms": 382.73268193006516,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 771.0784790106118,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          385.08352695498616,
          384.2154770391062
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 2402.611035038717,
      "sampled_peak_tree_pss_bytes": 45348864,
      "max_observed_processes": 1,
      "pss_samples": 208,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 659.2274410650134,
        "first_ms": 403.6085410043597,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 581.0709990328178,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          277.64342504087836,
          264.5031249849126
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 1888.1232819985598,
      "sampled_peak_tree_pss_bytes": 55153664,
      "max_observed_processes": 3,
      "pss_samples": 147,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 130.80310495570302,
        "first_ms": 4.555376013740897,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.006029995158315,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.977418993599713,
          2.172571956180036
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 314.3932190723717,
      "sampled_peak_tree_pss_bytes": 58788864,
      "max_observed_processes": 1,
      "pss_samples": 28,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 574.8138800263405,
        "first_ms": 968.8988199923187,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 55.832959013059735,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          6.828538957051933,
          10.709149995818734
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2144.250682089478,
      "sampled_peak_tree_pss_bytes": 173059072,
      "max_observed_processes": 1,
      "pss_samples": 159,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 376.15757808089256,
        "first_ms": 853.4453739412129,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.404610911384225,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          4.175631096586585,
          3.023028955794871
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1622.6216029608622,
      "sampled_peak_tree_pss_bytes": 180881408,
      "max_observed_processes": 1,
      "pss_samples": 123,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.9507,
        "first_ms": 4.388313,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 7.370519,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.930291,
          3.243253
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 144.79982294142246,
      "sampled_peak_tree_pss_bytes": 26618880,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 667.1763849444687,
        "first_ms": 4.1272989474236965,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 41.54512798413634,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          4.845967050641775,
          4.998457035981119
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 988.9814539346844,
      "sampled_peak_tree_pss_bytes": 85416960,
      "max_observed_processes": 1,
      "pss_samples": 84,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1095.2798869693652,
        "first_ms": 684.3571410281584,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1155.7762450538576,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          574.9454629840329,
          579.1680599795654
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3105.937825050205,
      "sampled_peak_tree_pss_bytes": 197106688,
      "max_observed_processes": 3,
      "pss_samples": 221,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 20.70911,
        "first_ms": 4.341174,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 8.155588,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.010303,
          4.13221
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 149.53473396599293,
      "sampled_peak_tree_pss_bytes": 34265088,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.899728,
        "first_ms": 2.783757,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.727721,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.25252,
          3.353683
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 145.25000704452395,
      "sampled_peak_tree_pss_bytes": 26858496,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1071.3364480761811,
        "first_ms": 579.3138660956174,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 846.4027609443292,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          421.8082780716941,
          422.7783619426191
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2695.786839001812,
      "sampled_peak_tree_pss_bytes": 73286656,
      "max_observed_processes": 4,
      "pss_samples": 201,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 23.226845,
        "first_ms": 5.138356,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 11.808664,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.174682,
          5.617126
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 148.45506695564836,
      "sampled_peak_tree_pss_bytes": 34249728,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 718.0852860910818,
        "first_ms": 737.8493059659377,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 38.91422809101641,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.9321700567379594,
          3.571758046746254
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1905.7905660010874,
      "sampled_peak_tree_pss_bytes": 173221888,
      "max_observed_processes": 1,
      "pss_samples": 147,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 21.353342,
        "first_ms": 3.185141,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.375649,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.44392,
          3.686517
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30068,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 147.20829098951072,
      "sampled_peak_tree_pss_bytes": 27267072,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/dense/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 20.99473,
        "first_ms": 4.108346,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 8.408076,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.134517,
          4.260145
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 147.8458819910884,
      "sampled_peak_tree_pss_bytes": 34179072,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-skew/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 674.8761229682714,
        "first_ms": 4.060929059050977,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 38.66010194178671,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.4238219959661365,
          3.286400926299393
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 955.1580570405349,
      "sampled_peak_tree_pss_bytes": 80243712,
      "max_observed_processes": 1,
      "pss_samples": 82,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 526.5073060290888,
        "first_ms": 1073.2313829939812,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.821127030998468,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.803742933087051,
          6.7648859694600105
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2160.754013922997,
      "sampled_peak_tree_pss_bytes": 180942848,
      "max_observed_processes": 1,
      "pss_samples": 158,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 577.5343640707433,
        "first_ms": 2.064041909761727,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.512001952156425,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.017379087395966,
          2.064612112008035
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 836.6717980243266,
      "sampled_peak_tree_pss_bytes": 79858688,
      "max_observed_processes": 1,
      "pss_samples": 72,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/dense/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 537.0376009959728,
        "first_ms": 650.4604850197211,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.819122014567256,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          6.173175992444158,
          6.349121918901801
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1760.8351520029828,
      "sampled_peak_tree_pss_bytes": 180162560,
      "max_observed_processes": 2,
      "pss_samples": 129,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:serial/sparse-uniform/1",
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 31.065334,
        "first_ms": 6.802417,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 7.93879,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.223793,
          5.546105
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 165.33582203555852,
      "sampled_peak_tree_pss_bytes": 26556416,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 126.73185695894063,
        "first_ms": 3.3875280059874058,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.620954002253711,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.035272005945444,
          3.744698013179004
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 295.2150630299002,
      "sampled_peak_tree_pss_bytes": 58786816,
      "max_observed_processes": 1,
      "pss_samples": 26,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/2",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 529.7976389992982,
        "first_ms": 4.116261028684676,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.05160799250007,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.146648989059031,
          2.6777059538289905
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 767.4080410506576,
      "sampled_peak_tree_pss_bytes": 85941248,
      "max_observed_processes": 1,
      "pss_samples": 66,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 22.077486,
        "first_ms": 3.341803,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.205032,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.98406,
          3.214132
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36936,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 149.36130796559155,
      "sampled_peak_tree_pss_bytes": 34407424,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-skew/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 792.3122049542144,
        "first_ms": 204.43116803653538,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 437.6935039181262,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          203.64444295410067,
          208.10058410279453
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1586.5930039435625,
      "sampled_peak_tree_pss_bytes": 37984256,
      "max_observed_processes": 1,
      "pss_samples": 137,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.812962,
        "first_ms": 2.828243,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 4.967837,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.367256,
          2.4051
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 136.7231080075726,
      "sampled_peak_tree_pss_bytes": 26639360,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 19.648456,
        "first_ms": 4.312391,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 9.124656,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.164745,
          5.947275
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 152.06425299402326,
      "sampled_peak_tree_pss_bytes": 34241536,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/dense/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 175.38766306824982,
        "first_ms": 6.049713934771717,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.604760001413524,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.5015709688887,
          3.362684976309538
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 379.88304893951863,
      "sampled_peak_tree_pss_bytes": 58828800,
      "max_observed_processes": 1,
      "pss_samples": 33,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 946.8370230169967,
        "first_ms": 995.265697943978,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 680.239328998141,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          328.21655506268144,
          316.6321750031784
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2864.7386230295524,
      "sampled_peak_tree_pss_bytes": 210283520,
      "max_observed_processes": 3,
      "pss_samples": 195,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 953.9683260954916,
        "first_ms": 424.9482969753444,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 534.6841380232945,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          246.73869996331632,
          249.12469799164683
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2158.5134750930592,
      "sampled_peak_tree_pss_bytes": 86201344,
      "max_observed_processes": 6,
      "pss_samples": 159,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/dense/1",
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 21.877067,
        "first_ms": 4.087333,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 7.73386,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.013065,
          3.708109
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 148.41373800300062,
      "sampled_peak_tree_pss_bytes": 34256896,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 32.154643,
        "first_ms": 10.256198,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 12.941379,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.9773,
          5.808496
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 161.49003093596548,
      "sampled_peak_tree_pss_bytes": 26675200,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 30.270552,
        "first_ms": 6.459839,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.327234,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.165372,
          5.071773
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 159.8129259655252,
      "sampled_peak_tree_pss_bytes": 26873856,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 537.1412520762533,
        "first_ms": 3.8143120473250747,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.0514729982242,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.8022410590201616,
          2.8642480028793216
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 778.7197030847892,
      "sampled_peak_tree_pss_bytes": 85934080,
      "max_observed_processes": 1,
      "pss_samples": 68,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 41.109673,
        "first_ms": 6.842842,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 10.275732,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.447888,
          7.63666
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 170.2748619718477,
      "sampled_peak_tree_pss_bytes": 26672128,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1101.8612780608237,
        "first_ms": 541.1062471102923,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 781.7991939373314,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          369.3472020095214,
          410.7696210267022
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2681.469160015695,
      "sampled_peak_tree_pss_bytes": 247489536,
      "max_observed_processes": 6,
      "pss_samples": 191,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 24.469054,
        "first_ms": 3.684858,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.393851,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          4.284149,
          4.098348
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37208,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 145.42544004507363,
      "sampled_peak_tree_pss_bytes": 34490368,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 20.568977,
        "first_ms": 2.91012,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.950846,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.293157,
          3.503853
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 142.13579101487994,
      "sampled_peak_tree_pss_bytes": 26805248,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 20.513848,
        "first_ms": 3.804893,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.421795,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.51864,
          2.786814
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30352,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 147.21053699031472,
      "sampled_peak_tree_pss_bytes": 27419648,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 42.636058,
        "first_ms": 5.441529,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 9.711417,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.825055,
          2.787741
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 180.8476879959926,
      "sampled_peak_tree_pss_bytes": 26813440,
      "max_observed_processes": 1,
      "pss_samples": 16,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-uniform/4",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 566.5239010704681,
        "first_ms": 3.6911360220983624,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.48576799221337,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.4572679540142417,
          2.8087550308555365
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 906.6860129823908,
      "sampled_peak_tree_pss_bytes": 105486336,
      "max_observed_processes": 1,
      "pss_samples": 77,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 30.074782,
        "first_ms": 4.047327,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.54292,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.186266,
          5.347243
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37620,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 162.18857490457594,
      "sampled_peak_tree_pss_bytes": 34989056,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/dense/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 23.478092,
        "first_ms": 3.643655,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.926263,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.32058,
          4.597287
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37784,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 148.7200940027833,
      "sampled_peak_tree_pss_bytes": 34979840,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.985658,
        "first_ms": 3.166349,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.010895,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.023677,
          2.882174
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 140.88875392917544,
      "sampled_peak_tree_pss_bytes": 26887168,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 41.125323,
        "first_ms": 7.37877,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.054756,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.07357,
          2.731347
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30248,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 172.7186730131507,
      "sampled_peak_tree_pss_bytes": 27309056,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/dense/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 42.630982,
        "first_ms": 6.988957,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 16.148369,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          8.347412,
          7.782205
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 182.05694598145783,
      "sampled_peak_tree_pss_bytes": 34274304,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 1762.3577299527824,
        "first_ms": 766.385650029406,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1009.867953020148,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          482.3647029697895,
          477.3852189537138
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3865.7290700357407,
      "sampled_peak_tree_pss_bytes": 65714176,
      "max_observed_processes": 4,
      "pss_samples": 288,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 1613.0910590291023,
        "first_ms": 398.2065470190719,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 416.79961106274277,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          188.07968101464212,
          178.83087904192507
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2698.8504680339247,
      "sampled_peak_tree_pss_bytes": 89328640,
      "max_observed_processes": 6,
      "pss_samples": 200,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-uniform/4",
        "language": "python",
        "backend": "process-copy",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 1641.4385269163176,
        "first_ms": 471.91847499925643,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 803.4812869736925,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          345.0211070012301,
          409.2541899299249
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 3205.522622098215,
      "sampled_peak_tree_pss_bytes": 181053440,
      "max_observed_processes": 6,
      "pss_samples": 231,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:serial/sparse-uniform/1",
        "language": "python",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 1729.3118350207806,
        "first_ms": 460.8336000237614,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 817.3025889554992,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          451.0316390078515,
          320.7492439541966
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 3218.1274120230228,
      "sampled_peak_tree_pss_bytes": 37983232,
      "max_observed_processes": 1,
      "pss_samples": 268,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/4",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 37.915625,
        "first_ms": 9.361603,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 9.519635,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.279906,
          3.012006
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 168.35557902231812,
      "sampled_peak_tree_pss_bytes": 26668032,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-skew/1",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 595.3007019124925,
        "first_ms": 559.7582310438156,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.671821096912026,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.8670369647443295,
          3.143746987916529
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1528.596154996194,
      "sampled_peak_tree_pss_bytes": 173073408,
      "max_observed_processes": 1,
      "pss_samples": 121,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 21.96128,
        "first_ms": 3.25685,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.624599,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.060628,
          4.556882
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37172,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 148.23812094982713,
      "sampled_peak_tree_pss_bytes": 34460672,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/4",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.386561,
        "first_ms": 2.413092,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.706758,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.748116,
          1.857677
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 135.02879498992115,
      "sampled_peak_tree_pss_bytes": 26896384,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/4",
        "language": "python",
        "backend": "process-shared",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 921.338074025698,
        "first_ms": 415.7098529394716,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 488.56372700538486,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          241.30997899919748,
          245.55357301142067
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2038.271842058748,
      "sampled_peak_tree_pss_bytes": 93978624,
      "max_observed_processes": 6,
      "pss_samples": 149,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/1",
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.734948,
        "first_ms": 2.26186,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.521242,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.555955,
          2.855211
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30168,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 139.89009300712496,
      "sampled_peak_tree_pss_bytes": 27321344,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/4",
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 38.334694,
        "first_ms": 2.654636,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 8.234688,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          5.135784,
          2.895503
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 164.33220298495144,
      "sampled_peak_tree_pss_bytes": 26646528,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 35.925268,
        "first_ms": 4.875527,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.507571,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          1.758965,
          2.6526
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 154.11194507032633,
      "sampled_peak_tree_pss_bytes": 26866688,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-skew/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 42.443888,
        "first_ms": 3.513565,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.381998,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          9.435327,
          3.83189
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 171.39813606627285,
      "sampled_peak_tree_pss_bytes": 26821632,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/1",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.518845,
        "first_ms": 2.5722,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.339331,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.392372,
          1.843487
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 134.5045689959079,
      "sampled_peak_tree_pss_bytes": 26822656,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.147553,
        "first_ms": 2.466046,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.285218,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.114779,
          3.059954
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 141.0497889155522,
      "sampled_peak_tree_pss_bytes": 26838016,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.774516,
        "first_ms": 4.39126,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.666093,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.94068,
          1.616233
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 138.52512906305492,
      "sampled_peak_tree_pss_bytes": 26833920,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-uniform/1",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.461577,
        "first_ms": 3.030424,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 6.164111,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.139918,
          2.830766
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 133.5329880239442,
      "sampled_peak_tree_pss_bytes": 26606592,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/dense/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 983.6381380446255,
        "first_ms": 449.88690107129514,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 762.2933640377596,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          381.87291100621223,
          378.61296895425767
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2391.823904006742,
      "sampled_peak_tree_pss_bytes": 62791680,
      "max_observed_processes": 3,
      "pss_samples": 181,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 816.6254879906774,
        "first_ms": 205.87505400180817,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 437.90383893065155,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          204.1392750106752,
          205.80097509082407
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1614.6275280043483,
      "sampled_peak_tree_pss_bytes": 37985280,
      "max_observed_processes": 1,
      "pss_samples": 138,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/sparse-uniform/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.327301,
        "first_ms": 7.215392,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.029137,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.94859,
          5.946731
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 151.6626119846478,
      "sampled_peak_tree_pss_bytes": 26878976,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-serial/dense/1",
        "language": "python",
        "backend": "numpy-serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 242.37284006085247,
        "first_ms": 6.633857032284141,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.190605910494924,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          5.315757938660681,
          6.215728004463017
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 470.84644495043904,
      "sampled_peak_tree_pss_bytes": 58749952,
      "max_observed_processes": 1,
      "pss_samples": 40,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-uniform/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 902.9838129645213,
        "first_ms": 605.8156529907137,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 49.632072914391756,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          6.668232032097876,
          5.983822979032993
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 2081.379645038396,
      "sampled_peak_tree_pss_bytes": 170808320,
      "max_observed_processes": 1,
      "pss_samples": 158,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 958.115701097995,
        "first_ms": 296.2162309559062,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 580.958595033735,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          276.21745807118714,
          276.7766669858247
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2020.9367730421945,
      "sampled_peak_tree_pss_bytes": 37983232,
      "max_observed_processes": 1,
      "pss_samples": 167,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-uniform/2",
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 33.483593,
        "first_ms": 4.864051,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.988476,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          1.752291,
          6.138901
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30056,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 164.02063297573477,
      "sampled_peak_tree_pss_bytes": 27358208,
      "max_observed_processes": 1,
      "pss_samples": 14,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/dense/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 507.2091210167855,
        "first_ms": 982.9291389323771,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.118851020000875,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          6.686232052743435,
          3.243932966142893
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 2022.3389350576326,
      "sampled_peak_tree_pss_bytes": 181020672,
      "max_observed_processes": 1,
      "pss_samples": 148,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-skew/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 37.017232,
        "first_ms": 5.350622,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 8.000969,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          5.787677,
          2.108312
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 167.95072599779814,
      "sampled_peak_tree_pss_bytes": 26808320,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-serial/sparse-skew/1",
        "language": "python",
        "backend": "numba-serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 563.0465489812195,
        "first_ms": 340.84188798442483,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.77175298798829,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.7173710986971855,
          2.0914619090035558
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 1244.4954719394445,
      "sampled_peak_tree_pss_bytes": 170871808,
      "max_observed_processes": 1,
      "pss_samples": 100,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/1",
        "language": "python",
        "backend": "process-copy",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 841.7755259433761,
        "first_ms": 411.3819480407983,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 583.797391038388,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          276.9448719918728,
          276.3155859429389
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2044.9431269662455,
      "sampled_peak_tree_pss_bytes": 210563072,
      "max_observed_processes": 3,
      "pss_samples": 148,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.83851,
        "first_ms": 2.250034,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.664942,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.87495,
          2.685379
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 135.8074969612062,
      "sampled_peak_tree_pss_bytes": 26866688,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numpy-thread/sparse-skew/1",
        "language": "python",
        "backend": "numpy-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 525.1886330079287,
        "first_ms": 3.22772900108248,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.73700196668506,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.9488109285011888,
          2.2372729144990444
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      },
      "process_wall_ms": 760.225472971797,
      "sampled_peak_tree_pss_bytes": 80626688,
      "max_observed_processes": 1,
      "pss_samples": 65,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/dense/1",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 20.801186,
        "first_ms": 3.535597,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.684564,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.501295,
          3.175101
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36928,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 147.69824198447168,
      "sampled_peak_tree_pss_bytes": 34427904,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/dense/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 1048.8793730037287,
        "first_ms": 627.1810010075569,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 854.7143259784207,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          439.76601399481297,
          413.1875380408019
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2730.2397809689865,
      "sampled_peak_tree_pss_bytes": 244898816,
      "max_observed_processes": 4,
      "pss_samples": 188,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/2",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 610.0627740379423,
        "first_ms": 582.2682139696553,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 36.0946599394083,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          2.561102039180696,
          3.4357589902356267
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1563.5177640942857,
      "sampled_peak_tree_pss_bytes": 173135872,
      "max_observed_processes": 1,
      "pss_samples": 123,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/sparse-uniform/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.459454,
        "first_ms": 2.53241,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.107252,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.013376,
          1.988839
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 135.53012406919152,
      "sampled_peak_tree_pss_bytes": 26805248,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:tbb/sparse-skew/4",
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 19.459092,
        "first_ms": 2.922309,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.871729,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.710707,
          2.048928
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30300,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 138.00374197307974,
      "sampled_peak_tree_pss_bytes": 27407360,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:std-thread/dense/2",
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 21.416551,
        "first_ms": 3.493307,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.731636,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.721366,
          4.002644
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37096,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 148.4293470857665,
      "sampled_peak_tree_pss_bytes": 34449408,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 789.7006130078807,
        "first_ms": 198.73255700804293,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 427.4304020218551,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          200.5269449437037,
          200.24007896427065
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1609.6716719912365,
      "sampled_peak_tree_pss_bytes": 37980160,
      "max_observed_processes": 1,
      "pss_samples": 136,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:numba-parallel/sparse-uniform/4",
        "language": "python",
        "backend": "numba-parallel",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 558.5937490686774,
        "first_ms": 557.8452249756083,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 37.287198938429356,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          3.0432110652327538,
          3.498784964904189
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      },
      "process_wall_ms": 1486.4624039037153,
      "sampled_peak_tree_pss_bytes": 173303808,
      "max_observed_processes": 1,
      "pss_samples": 117,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-skew/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.584711,
        "first_ms": 2.620595,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 6.300368,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.428377,
          3.67267
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 136.31818897556514,
      "sampled_peak_tree_pss_bytes": 26642432,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/1",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 22.439781,
        "first_ms": 3.26584,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 7.028416,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.913214,
          3.106471
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36928,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 147.7085630176589,
      "sampled_peak_tree_pss_bytes": 34426880,
      "max_observed_processes": 1,
      "pss_samples": 13,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-uniform/2",
        "language": "python",
        "backend": "process-shared",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 786.2839730223641,
        "first_ms": 308.0862230854109,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 344.29056104272604,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          154.42179108504206,
          172.86766308825463
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 1658.6330339778215,
      "sampled_peak_tree_pss_bytes": 65669120,
      "max_observed_processes": 4,
      "pss_samples": 127,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 1012.9516030428931,
        "first_ms": 271.6763069620356,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 555.8895600261167,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          262.79901002999395,
          265.24227496702224
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 2019.9780970579013,
      "sampled_peak_tree_pss_bytes": 37991424,
      "max_observed_processes": 1,
      "pss_samples": 167,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:rayon/sparse-uniform/2",
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 37.939479,
        "first_ms": 6.353355,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 9.183571,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          6.101731,
          2.865497
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 170.66347901709378,
      "sampled_peak_tree_pss_bytes": 26658816,
      "max_observed_processes": 1,
      "pss_samples": 15,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/4",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 17.852107,
        "first_ms": 2.577717,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 4.478377,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.385563,
          1.987904
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 137.40583509206772,
      "sampled_peak_tree_pss_bytes": 26833920,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/dense/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 20.220032,
        "first_ms": 3.450519,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.36408,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.988018,
          3.369116
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36928,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 141.13299804739654,
      "sampled_peak_tree_pss_bytes": 34476032,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-copy/sparse-skew/2",
        "language": "python",
        "backend": "process-copy",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 803.7442159838974,
        "first_ms": 439.87242493312806,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 575.3705190727487,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          284.01505295187235,
          261.1078249756247
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 2061.3953220890835,
      "sampled_peak_tree_pss_bytes": 225671168,
      "max_observed_processes": 4,
      "pss_samples": 150,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/dense/2",
        "language": "python",
        "backend": "thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 995.3264739597216,
        "first_ms": 333.8637870037928,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 683.5416089743376,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          339.9558949749917,
          341.7968839639798
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      },
      "process_wall_ms": 2175.330762984231,
      "sampled_peak_tree_pss_bytes": 45398016,
      "max_observed_processes": 1,
      "pss_samples": 181,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-skew/4",
        "language": "python",
        "backend": "thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 844.2442750092596,
        "first_ms": 210.7438490493223,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 439.7059800103307,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          200.037342030555,
          212.4668110627681
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1648.9462080644444,
      "sampled_peak_tree_pss_bytes": 37984256,
      "max_observed_processes": 1,
      "pss_samples": 135,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:serial/sparse-uniform/1",
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 20.234237,
        "first_ms": 2.728393,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 3.997327,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          1.972439,
          1.9113
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 143.7551019480452,
      "sampled_peak_tree_pss_bytes": 26788864,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-dynamic/dense/4",
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 2,
        "setup_ms": 22.25602,
        "first_ms": 4.498899,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 6.158026,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.063564,
          3.087035
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37048,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 139.51930310577154,
      "sampled_peak_tree_pss_bytes": 34456576,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "cpp:omp-static/sparse-skew/2",
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.936454,
        "first_ms": 2.447689,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 5.093099,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          2.943558,
          2.040501
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      },
      "process_wall_ms": 136.05844392441213,
      "sampled_peak_tree_pss_bytes": 26866688,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:process-shared/sparse-skew/1",
        "language": "python",
        "backend": "process-shared",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 814.9995239218697,
        "first_ms": 292.25886589847505,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 486.2274390179664,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          230.05458805710077,
          229.87643408123404
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      },
      "process_wall_ms": 1764.9461190449074,
      "sampled_peak_tree_pss_bytes": 55273472,
      "max_observed_processes": 3,
      "pss_samples": 137,
      "stderr": ""
    },
    {
      "case": {
        "id": "python:thread/sparse-uniform/1",
        "language": "python",
        "backend": "thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 810.6319799553603,
        "first_ms": 209.4792149728164,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 445.32342604361475,
        "iterations": 2,
        "validation_scope": "all",
        "samples_ms": [
          208.19958799984306,
          209.8631839035079
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      },
      "process_wall_ms": 1622.0661620609462,
      "sampled_peak_tree_pss_bytes": 37979136,
      "max_observed_processes": 1,
      "pss_samples": 139,
      "stderr": ""
    },
    {
      "case": {
        "id": "rust:std-scoped/sparse-skew/2",
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64
      },
      "pass": "memory",
      "launch": 0,
      "measurement": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 2,
        "setup_ms": 18.752761,
        "first_ms": 3.24368,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 6.146103,
        "iterations": 2,
        "validations": "all",
        "samples_ms": [
          3.412756,
          2.527178
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      },
      "process_wall_ms": 135.40493603795767,
      "sampled_peak_tree_pss_bytes": 26651648,
      "max_observed_processes": 1,
      "pss_samples": 12,
      "stderr": ""
    }
  ],
  "summary": [
    {
      "id": "cpp:serial/dense/1",
      "language": "cpp",
      "backend": "serial",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.173696,
        "q1": 3.8851690000000003,
        "q3": 6.492654999999999,
        "min": 3.467056,
        "max": 12.748951,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.196542,
        "q1": 3.8766084999999997,
        "q3": 5.6965755,
        "min": 3.556675,
        "max": 7.196609,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 23.976909,
        "q1": 23.279809999999998,
        "q3": 32.3071635,
        "min": 22.582711,
        "max": 40.637418,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 56.66372203268111,
        "q1": 54.53098600264639,
        "q3": 81.17872598813847,
        "min": 52.398249972611666,
        "max": 105.69372994359583,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.87890625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 243.4279563765418,
        "q1": 179.11985791891146,
        "q3": 253.08833419774646,
        "min": 114.81175946128117,
        "max": 262.7487120189511,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.582711,
        "first_ms": 3.556675,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.029589,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.995467,
          3.774871,
          4.022044,
          3.750032,
          3.467247
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36992,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:serial/sparse-uniform/1",
      "language": "cpp",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.601784,
        "q1": 1.935197,
        "q3": 3.0076855,
        "min": 1.911572,
        "max": 4.067532,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.372172,
        "q1": 3.1276595,
        "q3": 3.5506035000000002,
        "min": 2.883147,
        "max": 3.729035,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 18.525786,
        "q1": 18.456659000000002,
        "q3": 19.7214905,
        "min": 18.387532,
        "max": 20.917195,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 40.61139200348407,
        "q1": 39.734603953547776,
        "q3": 43.882788508199155,
        "min": 38.85781590361148,
        "max": 47.15418501291424,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5478515625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 410.7567330216785,
        "q1": 363.5637548505893,
        "q3": 420.0619876876782,
        "min": 316.3707766795001,
        "max": 429.3672423536779,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.525786,
        "first_ms": 2.883147,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.645043,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.931945,
          2.063055,
          2.601784,
          2.853964,
          1.924868
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:serial/sparse-skew/1",
      "language": "cpp",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.614178,
        "q1": 2.2346245,
        "q3": 3.15145,
        "min": 1.970618,
        "max": 4.069697,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.786158,
        "q1": 2.6586245,
        "q3": 4.0228235,
        "min": 2.531091,
        "max": 5.259489,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.214942,
        "q1": 19.1526375,
        "q3": 19.889203000000002,
        "min": 19.090333,
        "max": 20.563464,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 44.53786194790155,
        "q1": 41.99657094432041,
        "q3": 45.341120508965105,
        "min": 39.455279940739274,
        "max": 46.14437907002866,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.56640625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 351.7200376030927,
        "q1": 328.5350937527941,
        "q3": 389.30199694764906,
        "min": 305.35014990249556,
        "max": 426.8839562922055,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.214942,
        "first_ms": 2.531091,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.712785,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.036449,
          2.591479,
          1.970618,
          2.764303,
          2.088078
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/dense/1",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.767062,
        "q1": 3.3771645,
        "q3": 4.339055999999999,
        "min": 2.954317,
        "max": 5.194531,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.334211,
        "q1": 4.2388575,
        "q3": 5.5619025,
        "min": 3.143504,
        "max": 5.789594,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 23.292873,
        "q1": 22.309869,
        "q3": 26.2618185,
        "min": 21.326865,
        "max": 29.230764,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 54.4741420308128,
        "q1": 51.86976050026715,
        "q3": 59.42673102254048,
        "min": 49.265378969721496,
        "max": 64.37932001426816,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.892578125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 261.4271643528794,
        "q1": 245.69049297312114,
        "q3": 269.91977611706136,
        "min": 229.9538215933629,
        "max": 278.4123878812434,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.326865,
        "first_ms": 3.143504,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.958971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.17521,
          2.984984,
          2.954317,
          4.502902,
          3.3228
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37096,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-uniform/1",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.002325,
        "q1": 1.9506709999999998,
        "q3": 2.5603534999999997,
        "min": 1.69494,
        "max": 3.594265,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.067998,
        "q1": 2.726332,
        "q3": 3.1108575,
        "min": 2.384666,
        "max": 3.153717,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.383064,
        "q1": 18.315662,
        "q3": 20.4256235,
        "min": 16.24826,
        "max": 20.468183,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 41.25466605182737,
        "q1": 39.21805904246867,
        "q3": 41.49994399631396,
        "min": 37.18145203310996,
        "max": 41.74522194080055,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.580078125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 422.92192869654696,
        "q1": 417.03484521640047,
        "q3": 439.2470801102709,
        "min": 411.147761736254,
        "max": 455.5722315239948,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.468183,
        "first_ms": 3.153717,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.161078,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.22138,
          2.968359,
          1.96417,
          1.968915,
          2.747549
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-skew/1",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.24839,
        "q1": 2.6912525,
        "q3": 3.5281725,
        "min": 2.150213,
        "max": 6.880208,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.948739,
        "q1": 2.6342225,
        "q3": 6.469292000000001,
        "min": 2.319706,
        "max": 9.989845,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 34.570393,
        "q1": 26.731961000000002,
        "q3": 40.4442835,
        "min": 18.893529,
        "max": 46.318174,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 58.53338190354407,
        "q1": 51.936978416051716,
        "q3": 74.56326199462637,
        "min": 45.34057492855936,
        "max": 90.59314208570868,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.62890625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 303.3481015323083,
        "q1": 247.66106042179462,
        "q3": 352.17006185791433,
        "min": 191.97401931128093,
        "max": 400.9920221835203,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.893529,
        "first_ms": 2.948739,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.482714,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.377077,
          3.289989,
          2.74601,
          3.458294,
          3.24839
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/dense/2",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.068039,
        "q1": 3.4638780000000002,
        "q3": 6.116175,
        "min": 3.032684,
        "max": 8.52006,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.032746,
        "q1": 4.3354365,
        "q3": 7.263885500000001,
        "min": 3.638127,
        "max": 9.495025,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 25.146657,
        "q1": 23.6645375,
        "q3": 36.2202595,
        "min": 22.182418,
        "max": 47.293862,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 62.08768906071782,
        "q1": 58.25611302861944,
        "q3": 77.97688053688034,
        "min": 54.424536996521056,
        "max": 93.86607201304287,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.853515625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 205.53350455096395,
        "q1": 188.03001385034491,
        "q3": 229.93952452390067,
        "min": 170.52652314972585,
        "max": 254.3455444968374,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 47.293862,
        "first_ms": 9.495025,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 29.320952,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.09537,
          7.723671,
          7.432659,
          3.032684,
          4.018089
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37080,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-uniform/2",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.202057,
        "q1": 2.007313,
        "q3": 2.5062290000000003,
        "min": 1.919407,
        "max": 2.941309,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.834481,
        "q1": 2.5612095,
        "q3": 3.10294,
        "min": 2.287938,
        "max": 3.371399,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.470517,
        "q1": 19.061291500000003,
        "q3": 19.851905000000002,
        "min": 18.652066,
        "max": 20.233293,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 39.90004200022668,
        "q1": 39.610846957657486,
        "q3": 40.972426533699036,
        "min": 39.321651915088296,
        "max": 42.044811067171395,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6337890625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 434.9871282958866,
        "q1": 415.30541690756104,
        "q3": 437.494345590719,
        "min": 395.6237055192355,
        "max": 440.0015628855514,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.652066,
        "first_ms": 3.371399,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.494593,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.005091,
          2.481766,
          2.530692,
          2.184873,
          2.00667
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-skew/2",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.747665,
        "q1": 2.190415,
        "q3": 3.2340285,
        "min": 2.030335,
        "max": 3.724955,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.937845,
        "q1": 2.6329995,
        "q3": 2.9861405,
        "min": 2.328154,
        "max": 3.034436,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 18.754677,
        "q1": 18.713783499999998,
        "q3": 18.897891,
        "min": 18.67289,
        "max": 19.041105,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 42.20401495695114,
        "q1": 41.20289796264842,
        "q3": 43.49293798441067,
        "min": 40.2017809683457,
        "max": 44.781861011870205,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5791015625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 356.32882394455584,
        "q1": 333.2804651497837,
        "q3": 387.5149271210906,
        "min": 310.23210635501164,
        "max": 418.7010302976253,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.041105,
        "first_ms": 2.328154,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.031983,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.106293,
          2.543009,
          2.267813,
          2.113017,
          3.724955
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/dense/4",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.116798,
        "q1": 3.843991,
        "q3": 4.19904,
        "min": 2.431078,
        "max": 5.169271,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.333341,
        "q1": 4.3209975,
        "q3": 4.3562655,
        "min": 4.308654,
        "max": 4.37919,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 24.093664,
        "q1": 23.6339575,
        "q3": 24.676698000000002,
        "min": 23.174251,
        "max": 25.259732,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 55.759264971129596,
        "q1": 53.52495750412345,
        "q3": 57.280168461147696,
        "min": 51.2906500371173,
        "max": 58.801071951165795,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.8642578125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 246.96837673662606,
        "q1": 238.43794293266677,
        "q3": 264.0285808887878,
        "min": 229.90750912870752,
        "max": 281.08878504094963,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 25.259732,
        "first_ms": 4.308654,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.747876,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.172229,
          4.095154,
          4.11599,
          4.168486,
          5.169271
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37208,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-uniform/4",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.010935,
        "q1": 2.1246105,
        "q3": 4.0700595,
        "min": 1.726217,
        "max": 6.046009,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.717312,
        "q1": 2.6218245,
        "q3": 2.972047,
        "min": 2.526337,
        "max": 3.226782,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 22.278145,
        "q1": 21.3257865,
        "q3": 26.929094499999998,
        "min": 20.373428,
        "max": 31.580044,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 53.00519394222647,
        "q1": 47.96924092806876,
        "q3": 55.25150347966701,
        "min": 42.933287913911045,
        "max": 57.497813017107546,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.650390625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 306.7798406131474,
        "q1": 279.72072692271007,
        "q3": 340.4089447794344,
        "min": 252.6616132322727,
        "max": 374.0380489457214,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 22.278145,
        "first_ms": 3.226782,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.789314,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.217639,
          3.613169,
          2.758379,
          4.199035,
          4.60954
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:std-thread/sparse-skew/4",
      "language": "cpp",
      "backend": "std-thread",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 5.901932,
        "q1": 5.174151,
        "q3": 6.8614785000000005,
        "min": 2.183137,
        "max": 9.953164,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.873101,
        "q1": 3.1930205000000003,
        "q3": 6.395995,
        "min": 2.51294,
        "max": 8.918889,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 25.562233,
        "q1": 22.958759,
        "q3": 31.2021785,
        "min": 20.355285,
        "max": 36.842124,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 65.38574700243771,
        "q1": 63.34614002844319,
        "q3": 79.00533947395161,
        "min": 61.306533054448664,
        "max": 92.6249319454655,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.65234375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 180.25223560538993,
        "q1": 156.00184535069894,
        "q3": 183.84356973164006,
        "min": 131.75145509600793,
        "max": 187.43490385789016,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "std-thread",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.355285,
        "first_ms": 2.51294,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 27.738907,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.183137,
          5.050932,
          5.901932,
          6.820587,
          7.421495
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/dense/1",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.827015,
        "q1": 3.2093350000000003,
        "q3": 6.398012,
        "min": 3.049178,
        "max": 10.090424,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.84972,
        "q1": 3.5258215,
        "q3": 5.3813745,
        "min": 3.201923,
        "max": 6.913029,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 36.637294,
        "q1": 29.494070999999998,
        "q3": 39.2758475,
        "min": 22.350848,
        "max": 41.914401,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 72.147605009377,
        "q1": 61.15773500641808,
        "q3": 83.35982047719881,
        "min": 50.167865003459156,
        "max": 94.57203594502062,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.8330078125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 272.2442915544703,
        "q1": 204.2190397035751,
        "q3": 282.34316586771087,
        "min": 136.19378785267995,
        "max": 292.44204018095144,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 41.914401,
        "first_ms": 3.201923,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.365858,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.401213,
          3.225088,
          3.109995,
          4.416333,
          3.193582
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36948,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-uniform/1",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.592792,
        "q1": 1.9511405,
        "q3": 3.3829755,
        "min": 1.79844,
        "max": 6.181124,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.790477,
        "q1": 2.379966,
        "q3": 3.8227885,
        "min": 1.969455,
        "max": 4.8551,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.360563,
        "q1": 19.783824,
        "q3": 28.1385485,
        "min": 19.207085,
        "max": 35.916534,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 41.86070105060935,
        "q1": 40.67236353876069,
        "q3": 59.39077807124704,
        "min": 39.484026026912034,
        "max": 76.92085509188473,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5478515625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 372.86325120945656,
        "q1": 311.6057804063037,
        "q3": 414.089612279451,
        "min": 250.34830960315088,
        "max": 455.31597334944547,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.207085,
        "first_ms": 2.790477,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.409742,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.44302,
          2.456822,
          1.979186,
          1.923095,
          3.322931
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-skew/1",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.337705,
        "q1": 2.4352085,
        "q3": 5.451033,
        "min": 1.929551,
        "max": 10.663085,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.587126,
        "q1": 2.855269,
        "q3": 4.5114835,
        "min": 2.123412,
        "max": 5.435841,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.744833,
        "q1": 19.014075,
        "q3": 24.972109,
        "min": 18.283317,
        "max": 30.199385,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 45.877494965679944,
        "q1": 42.63029003050178,
        "q3": 64.91815345361829,
        "min": 39.38308509532362,
        "max": 83.95881194155663,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6142578125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 314.2252537557459,
        "q1": 226.246663339385,
        "q3": 370.95862491254263,
        "min": 138.2680729230241,
        "max": 427.69199606933944,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.744833,
        "first_ms": 3.587126,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.912152,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.93071,
          3.401407,
          3.337705,
          2.711105,
          3.231658
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/dense/2",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.811251,
        "q1": 3.5657095,
        "q3": 4.4394255000000005,
        "min": 2.979439,
        "max": 6.095793,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.680213,
        "q1": 3.3987230000000004,
        "q3": 3.9853,
        "min": 3.117233,
        "max": 4.290387,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 25.255626,
        "q1": 23.5822805,
        "q3": 25.9078065,
        "min": 21.908935,
        "max": 26.559987,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 57.561296969652176,
        "q1": 53.94053302006796,
        "q3": 59.537908993661404,
        "min": 50.319769070483744,
        "max": 61.51452101767063,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.8408203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 243.4874888336638,
        "q1": 228.0409119247131,
        "q3": 265.20636511623854,
        "min": 212.5943350157624,
        "max": 286.92524139881334,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 26.559987,
        "first_ms": 4.290387,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 23.518971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.095793,
          4.058438,
          5.383302,
          3.518008,
          4.436338
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36936,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-uniform/2",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.25664,
        "q1": 2.6594955000000002,
        "q3": 3.5184825,
        "min": 1.768919,
        "max": 5.026683,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.971034,
        "q1": 3.4760935,
        "q3": 4.3271115,
        "min": 2.981153,
        "max": 4.683189,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 22.017193,
        "q1": 18.567332999999998,
        "q3": 25.233781999999998,
        "min": 15.117473,
        "max": 28.450371,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 51.35295097716153,
        "q1": 43.10497496044263,
        "q3": 56.56572349835187,
        "min": 34.85699894372374,
        "max": 61.77849601954222,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5712890625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 301.74385012841617,
        "q1": 276.8039796834588,
        "q3": 350.2600428084745,
        "min": 251.86410923850144,
        "max": 398.7762354885328,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 22.017193,
        "first_ms": 4.683189,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.570346,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.773494,
          3.481197,
          3.555768,
          2.905671,
          3.434701
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-skew/2",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.872967,
        "q1": 2.2373995,
        "q3": 5.9488825,
        "min": 2.01321,
        "max": 8.073309,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.091262,
        "q1": 2.6605315000000003,
        "q3": 4.810035,
        "min": 2.229801,
        "max": 6.528808,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.604195,
        "q1": 19.4175955,
        "q3": 30.624950499999997,
        "min": 19.230996,
        "max": 41.645706,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 43.04040200076997,
        "q1": 41.94356297375634,
        "q3": 68.17490950925276,
        "min": 40.84672394674271,
        "max": 93.30941701773554,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6220703125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 385.0670682864847,
        "q1": 265.70670019686423,
        "q3": 395.15014121631253,
        "min": 146.34633210724377,
        "max": 405.23321414614037,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 41.645706,
        "first_ms": 6.528808,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 34.16553,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.938356,
          6.85044,
          5.959409,
          6.867758,
          8.073309
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/dense/4",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.018938,
        "q1": 3.4544734999999998,
        "q3": 4.801958,
        "min": 2.969172,
        "max": 10.883932,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.489325,
        "q1": 3.4162220000000003,
        "q3": 5.910766499999999,
        "min": 3.343119,
        "max": 8.332208,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 22.221695,
        "q1": 22.137779000000002,
        "q3": 32.3452355,
        "min": 22.053863,
        "max": 42.468776,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 53.185936994850636,
        "q1": 51.59954144619405,
        "q3": 74.746033991687,
        "min": 50.01314589753747,
        "max": 96.30613098852336,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.9072265625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 243.06261618696885,
        "q1": 200.41155689567205,
        "q3": 266.7180763521724,
        "min": 157.76049760437527,
        "max": 290.37353651737595,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 22.221695,
        "first_ms": 3.489325,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.2192,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.02234,
          4.018938,
          3.038431,
          2.969172,
          4.152512
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37104,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-uniform/4",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.595633,
        "q1": 2.1875869999999997,
        "q3": 3.217368,
        "min": 1.930989,
        "max": 6.945264,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.909123,
        "q1": 2.549135,
        "q3": 3.2996715,
        "min": 2.189147,
        "max": 3.69022,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.947815,
        "q1": 20.152561499999997,
        "q3": 28.8644605,
        "min": 19.357308,
        "max": 36.781106,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 48.2963080285117,
        "q1": 45.93169753206894,
        "q3": 58.3341455203481,
        "min": 43.56708703562617,
        "max": 68.3719830121845,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6416015625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 321.66324084593055,
        "q1": 299.8310756563003,
        "q3": 331.37006691731216,
        "min": 277.9989104666701,
        "max": 341.07689298869377,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 36.781106,
        "first_ms": 2.909123,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.985682,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.792391,
          2.091997,
          1.930989,
          6.945264,
          1.945089
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-static/sparse-skew/4",
      "language": "cpp",
      "backend": "omp-static",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.442144,
        "q1": 2.1110495,
        "q3": 3.4418805,
        "min": 1.691754,
        "max": 5.050036,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.372213,
        "q1": 3.8630400000000003,
        "q3": 8.828323000000001,
        "min": 3.353867,
        "max": 13.284433,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 35.498224,
        "q1": 27.728226499999998,
        "q3": 37.384977,
        "min": 19.958229,
        "max": 39.27173,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 70.49792201723903,
        "q1": 57.11436952697113,
        "q3": 71.70375599525869,
        "min": 43.73081703670323,
        "max": 72.90958997327834,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5908203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 379.2784697418464,
        "q1": 321.48103315086234,
        "q3": 429.1005656494915,
        "min": 263.68359655987837,
        "max": 478.9226615571366,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-static",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 35.498224,
        "first_ms": 3.353867,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.96212,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.339167,
          3.544594,
          4.527015,
          2.237321,
          5.050036
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/dense/1",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.902926,
        "q1": 3.180926,
        "q3": 6.3418465,
        "min": 2.93908,
        "max": 10.751074,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.27406,
        "q1": 3.9655415000000005,
        "q3": 4.9449445,
        "min": 3.657023,
        "max": 5.615829,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 23.2987,
        "q1": 22.3170285,
        "q3": 35.7878885,
        "min": 21.335357,
        "max": 48.277077,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 53.688714979216456,
        "q1": 51.4469490153715,
        "q3": 83.27669353457168,
        "min": 49.20518305152655,
        "max": 112.8646720899269,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.83203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 281.0316875593117,
        "q1": 200.68586210888753,
        "q3": 282.15541243617906,
        "min": 120.34003665846332,
        "max": 283.27913731304636,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 48.277077,
        "first_ms": 5.615829,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 41.548932,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.219803,
          6.46389,
          9.787227,
          10.751074,
          8.297544
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36960,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-uniform/1",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.960401,
        "q1": 2.109789,
        "q3": 6.2170555,
        "min": 1.614039,
        "max": 9.98002,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.063247,
        "q1": 3.5897419999999998,
        "q3": 6.2220495,
        "min": 2.116237,
        "max": 7.380852,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 40.951848,
        "q1": 30.996851,
        "q3": 44.686578499999996,
        "min": 21.041854,
        "max": 48.421309,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 88.46253098454326,
        "q1": 65.78854448162019,
        "q3": 89.73384444834664,
        "min": 43.11455797869712,
        "max": 91.00515791215003,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5634765625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 230.0847604647381,
        "q1": 188.89378629962835,
        "q3": 297.228525063504,
        "min": 147.7028121345186,
        "max": 364.37228966226985,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 40.951848,
        "first_ms": 7.380852,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 33.851759,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.37275,
          6.749779,
          4.960401,
          6.539766,
          9.98002
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-skew/1",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.342108,
        "q1": 2.007019,
        "q3": 3.090001,
        "min": 1.959969,
        "max": 3.624792,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.014204,
        "q1": 2.9739475,
        "q3": 3.150448,
        "min": 2.933691,
        "max": 3.286692,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.094705,
        "q1": 18.917558,
        "q3": 19.391235,
        "min": 18.740411,
        "max": 19.687765,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 40.5922329518944,
        "q1": 40.31228949315846,
        "q3": 45.697588473558426,
        "min": 40.03234603442252,
        "max": 50.80294399522245,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5634765625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 401.758028852976,
        "q1": 370.43472962063333,
        "q3": 404.8545132159342,
        "min": 339.1114303882907,
        "max": 407.95099757889244,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.094705,
        "first_ms": 3.286692,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.256374,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.508172,
          2.003673,
          2.010365,
          1.997516,
          3.471737
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/dense/2",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.321049,
        "q1": 4.031871,
        "q3": 4.7429369999999995,
        "min": 3.365843,
        "max": 6.37083,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.290087,
        "q1": 3.9643495,
        "q3": 4.3290625,
        "min": 3.638612,
        "max": 4.368038,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 28.318438,
        "q1": 27.7889475,
        "q3": 28.6000815,
        "min": 27.259457,
        "max": 28.881725,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 63.13213403336704,
        "q1": 62.91974551277235,
        "q3": 65.14553702436388,
        "min": 62.707356992177665,
        "max": 67.15894001536071,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.8134765625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 235.11869144222044,
        "q1": 212.63590497953436,
        "q3": 241.851915913972,
        "min": 190.15311851684825,
        "max": 248.58514038572358,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.318438,
        "first_ms": 4.368038,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.265855,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.261258,
          4.162209,
          3.523638,
          3.973032,
          4.321049
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 36976,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-uniform/2",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.584867,
        "q1": 2.1076495,
        "q3": 2.8463155,
        "min": 2.016168,
        "max": 2.954457,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.774032,
        "q1": 2.502071,
        "q3": 2.8407595,
        "min": 2.23011,
        "max": 2.907487,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 21.100439,
        "q1": 20.39322,
        "q3": 21.114571,
        "min": 19.686001,
        "max": 21.128703,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 42.709890054538846,
        "q1": 41.59371752757579,
        "q3": 43.034833564888686,
        "min": 40.477545000612736,
        "max": 43.359777075238526,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6220703125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 394.7814316290672,
        "q1": 390.88040687728346,
        "q3": 399.8914377296072,
        "min": 386.97938212549974,
        "max": 405.0014438301473,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 21.100439,
        "first_ms": 2.774032,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.920585,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.93695,
          2.109111,
          2.043865,
          2.642339,
          2.907724
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-skew/2",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.656,
        "q1": 1.9949279999999998,
        "q3": 4.026455,
        "min": 1.63762,
        "max": 6.272844,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.19335,
        "q1": 4.0656145,
        "q3": 5.970356,
        "min": 2.937879,
        "max": 6.747362,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 28.863038,
        "q1": 24.0642025,
        "q3": 32.675165,
        "min": 19.265367,
        "max": 36.487292,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 59.685845975764096,
        "q1": 50.28475949075073,
        "q3": 66.23873149510473,
        "min": 40.883673005737364,
        "max": 72.79161701444536,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.6220703125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 278.01133819200777,
        "q1": 265.83541649579115,
        "q3": 341.1831856460532,
        "min": 253.65949479957453,
        "max": 404.35503310009864,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.265367,
        "first_ms": 2.937879,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 12.365371,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.0803,
          3.022299,
          3.296604,
          2.034916,
          1.640973
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/dense/4",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.786141,
        "q1": 2.7577350000000003,
        "q3": 4.413538,
        "min": 2.388089,
        "max": 7.952576,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.078723,
        "q1": 3.438054,
        "q3": 4.133624,
        "min": 2.797385,
        "max": 4.188525,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 28.087558,
        "q1": 24.646297,
        "q3": 31.3257465,
        "min": 21.205036,
        "max": 34.563935,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 60.5643599992618,
        "q1": 55.26451696641743,
        "q3": 66.19580351980403,
        "min": 49.96467393357307,
        "max": 71.82724704034626,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.8603515625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 262.3608424973352,
        "q1": 233.23203421931333,
        "q3": 275.649174968738,
        "min": 204.1032259412914,
        "max": 288.9375074401408,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 21.205036,
        "first_ms": 2.797385,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.30478,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          5.773244,
          3.000686,
          3.493956,
          2.494493,
          2.514241
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37024,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-uniform/4",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.192214,
        "q1": 2.096743,
        "q3": 3.199101,
        "min": 1.908172,
        "max": 6.533361,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.922732,
        "q1": 2.6568465,
        "q3": 3.3066234999999997,
        "min": 2.390961,
        "max": 3.690515,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.249623,
        "q1": 19.8684505,
        "q3": 20.385027,
        "min": 19.487278,
        "max": 20.520431,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 44.74819405004382,
        "q1": 44.36196951428428,
        "q3": 45.347956533078104,
        "min": 43.975744978524745,
        "max": 45.94771901611239,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5908203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 333.7115397450444,
        "q1": 325.0150389135222,
        "q3": 365.7729918232445,
        "min": 316.31853808200003,
        "max": 397.8344439014446,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.487278,
        "first_ms": 2.390961,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 14.983,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.063099,
          2.010234,
          6.533361,
          2.071114,
          1.908172
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:omp-dynamic/sparse-skew/4",
      "language": "cpp",
      "backend": "omp-dynamic",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.788435,
        "q1": 3.239986,
        "q3": 4.457782,
        "min": 1.905671,
        "max": 9.976504,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.144748,
        "q1": 3.836042,
        "q3": 5.2607135,
        "min": 3.527336,
        "max": 6.376679,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.010453,
        "q1": 19.8819295,
        "q3": 27.8772545,
        "min": 19.753406,
        "max": 35.744056,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 57.582305977120996,
        "q1": 56.8430470302701,
        "q3": 63.28772596316412,
        "min": 56.103788083419204,
        "max": 68.99314594920725,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.5947265625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 237.0346659880522,
        "q1": 220.42536938840237,
        "q3": 243.55386866975056,
        "min": 203.81607278875254,
        "max": 250.0730713514489,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "omp-dynamic",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.010453,
        "first_ms": 3.527336,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.531922,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.246803,
          3.788435,
          3.233169,
          9.976504,
          3.927476
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 29760,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/dense/1",
      "language": "cpp",
      "backend": "tbb",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.859281,
        "q1": 3.156918,
        "q3": 5.0467635,
        "min": 3.02996,
        "max": 5.729275,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.911714,
        "q1": 3.7330425,
        "q3": 5.677783,
        "min": 3.554371,
        "max": 7.443852,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 23.526583,
        "q1": 23.1556495,
        "q3": 26.476913,
        "min": 22.784716,
        "max": 29.427243,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 52.18559899367392,
        "q1": 51.911565475165844,
        "q3": 61.41522800317034,
        "min": 51.63753195665777,
        "max": 70.64485701266676,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 33.2900390625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 267.70550680935725,
        "q1": 231.4746831949867,
        "q3": 276.672926239939,
        "min": 195.24385958061617,
        "max": 285.6403456705208,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.526583,
        "first_ms": 3.554371,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.67724,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.857221,
          3.225735,
          3.058268,
          4.48428,
          3.02996
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37652,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-uniform/1",
      "language": "cpp",
      "backend": "tbb",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.697321,
        "q1": 2.5933245,
        "q3": 3.2306850000000003,
        "min": 1.941702,
        "max": 8.168071,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.581156,
        "q1": 2.2858739999999997,
        "q3": 3.271121,
        "min": 1.990592,
        "max": 3.961086,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.526585,
        "q1": 20.063086,
        "q3": 32.6207495,
        "min": 19.599587,
        "max": 44.714914,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 49.18149102013558,
        "q1": 45.145216514356434,
        "q3": 63.72201151680201,
        "min": 41.10894200857729,
        "max": 78.26253201346844,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.00390625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 343.35942864991074,
        "q1": 278.9819490098791,
        "q3": 382.46394125694036,
        "min": 214.60446936984744,
        "max": 421.56845386397,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.599587,
        "first_ms": 2.581156,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.86047,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          1.941702,
          2.97445,
          1.96461,
          2.697321,
          1.994488
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30120,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-skew/1",
      "language": "cpp",
      "backend": "tbb",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.3988,
        "q1": 2.0970505,
        "q3": 4.3639265,
        "min": 1.957106,
        "max": 6.178373,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.831088,
        "q1": 2.5771625,
        "q3": 4.395219,
        "min": 2.323237,
        "max": 5.95935,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.441243,
        "q1": 19.2520855,
        "q3": 27.9073755,
        "min": 19.062928,
        "max": 36.373508,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 42.10106306709349,
        "q1": 41.465130052529275,
        "q3": 59.120666002854705,
        "min": 40.82919703796506,
        "max": 76.14026893861592,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.0556640625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 367.76907162140856,
        "q1": 287.0892442902219,
        "q3": 394.1253490019866,
        "min": 206.40941695903533,
        "max": 420.4816263825646,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.062928,
        "first_ms": 2.831088,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 11.891126,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.054686,
          2.903902,
          2.091994,
          2.3988,
          2.16304
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30104,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/dense/2",
      "language": "cpp",
      "backend": "tbb",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.861293,
        "q1": 3.3217865,
        "q3": 4.079445,
        "min": 3.112608,
        "max": 6.970951,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.034482,
        "q1": 4.7214145,
        "q3": 5.1355125,
        "min": 4.408347,
        "max": 5.236543,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 23.812838,
        "q1": 23.364554,
        "q3": 25.1440135,
        "min": 22.91627,
        "max": 26.475189,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 54.42363198380917,
        "q1": 53.932975977659225,
        "q3": 61.480661504901946,
        "min": 53.44231997150928,
        "max": 68.53769102599472,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 33.359375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 271.9558970209625,
        "q1": 244.790536368485,
        "q3": 274.4222312766158,
        "min": 217.62517571600748,
        "max": 276.888565532269,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 23.812838,
        "first_ms": 4.408347,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.057806,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.305295,
          3.861293,
          3.946595,
          3.25047,
          3.671621
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37652,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-uniform/2",
      "language": "cpp",
      "backend": "tbb",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.728651,
        "q1": 2.477858,
        "q3": 3.391813,
        "min": 1.923454,
        "max": 7.758333,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.226498,
        "q1": 2.6632879999999997,
        "q3": 3.4429309999999997,
        "min": 2.100078,
        "max": 3.659364,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.360516,
        "q1": 19.3497785,
        "q3": 30.993186,
        "min": 18.339041,
        "max": 41.625856,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 46.843942953273654,
        "q1": 43.34201547317207,
        "q3": 63.48323851125315,
        "min": 39.84008799307048,
        "max": 80.12253406923264,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.0908203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 317.956511307392,
        "q1": 273.02069330806387,
        "q3": 361.8754654924252,
        "min": 228.0848753087357,
        "max": 405.79441967745834,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 41.625856,
        "first_ms": 2.100078,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 21.921664,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          7.758333,
          2.355584,
          4.420653,
          5.186846,
          1.923454
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30224,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-skew/2",
      "language": "cpp",
      "backend": "tbb",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.108502,
        "q1": 2.6888259999999997,
        "q3": 3.7665845,
        "min": 2.074886,
        "max": 6.511587,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.115674,
        "q1": 2.76374,
        "q3": 4.045071,
        "min": 2.411806,
        "max": 4.974468,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.009136,
        "q1": 19.6993985,
        "q3": 22.192965,
        "min": 19.389661,
        "max": 24.376794,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 45.42700503952801,
        "q1": 43.23859850410372,
        "q3": 52.9327315161936,
        "min": 41.05019196867943,
        "max": 60.438457992859185,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.0439453125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 315.2022485771928,
        "q1": 269.09978989486615,
        "q3": 354.18730306925727,
        "min": 222.99733121253948,
        "max": 393.1723575613217,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.389661,
        "first_ms": 3.115674,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.862831,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.678003,
          3.772105,
          2.699649,
          2.890702,
          3.496789
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30112,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/dense/4",
      "language": "cpp",
      "backend": "tbb",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.069478,
        "q1": 3.396021,
        "q3": 5.330595,
        "min": 3.212502,
        "max": 5.882213,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.704892,
        "q1": 3.6881085000000002,
        "q3": 5.563904,
        "min": 3.671325,
        "max": 7.422916,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 24.351409,
        "q1": 23.8529665,
        "q3": 27.360663000000002,
        "min": 23.354524,
        "max": 30.369917,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 57.483099051751196,
        "q1": 56.221964536234736,
        "q3": 67.33520951820537,
        "min": 54.96083002071828,
        "max": 77.18731998465955,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 33.3681640625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 257.06860250978133,
        "q1": 217.1834073137592,
        "q3": 263.6786188985273,
        "min": 177.29821211773708,
        "max": 270.2886352872733,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 24.351409,
        "first_ms": 3.704892,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.450061,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.746289,
          3.21387,
          3.283384,
          4.889732,
          3.293874
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "peak_rss_kib": 37784,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-uniform/4",
      "language": "cpp",
      "backend": "tbb",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.946701,
        "q1": 2.658798,
        "q3": 3.1614435,
        "min": 2.035577,
        "max": 3.983418,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.051744,
        "q1": 2.9407759999999996,
        "q3": 3.1499385,
        "min": 2.829808,
        "max": 3.248133,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.123674,
        "q1": 20.0141315,
        "q3": 20.1343135,
        "min": 19.904589,
        "max": 20.144953,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 45.50786700565368,
        "q1": 44.020435481797904,
        "q3": 46.576184045989066,
        "min": 42.53300395794213,
        "max": 47.64450108632445,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.1494140625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 326.14730795402835,
        "q1": 315.1679529290662,
        "q3": 354.0847336892995,
        "min": 304.18859790410403,
        "max": 382.0221594245707,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.904589,
        "first_ms": 2.829808,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 13.088246,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.946701,
          2.035577,
          3.005049,
          2.091393,
          2.727786
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30208,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "cpp:tbb/sparse-skew/4",
      "language": "cpp",
      "backend": "tbb",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.066464,
        "q1": 2.845866,
        "q3": 4.432320000000001,
        "min": 2.184681,
        "max": 9.8033,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.613699,
        "q1": 3.2130615000000002,
        "q3": 7.282766499999999,
        "min": 2.812424,
        "max": 10.951834,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 21.61892,
        "q1": 21.295068999999998,
        "q3": 31.775348,
        "min": 20.971218,
        "max": 41.931776,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 52.437462960369885,
        "q1": 50.88129598880187,
        "q3": 74.25459497608244,
        "min": 49.32512901723385,
        "max": 96.071726991795,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 26.1376953125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 282.5157393756142,
        "q1": 229.63125814162595,
        "q3": 297.98030948256843,
        "min": 176.74677690763767,
        "max": 313.44487958952266,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "cpp",
        "backend": "tbb",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 20.971218,
        "first_ms": 2.812424,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 15.951768,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.980886,
          3.688966,
          2.488766,
          2.184681,
          4.289958
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "peak_rss_kib": 30200,
        "compiler": "13.3.0",
        "openmp": 201511,
        "tbb_version": "2021.11"
      }
    },
    {
      "id": "rust:serial/dense/1",
      "language": "rust",
      "backend": "serial",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.387059,
        "q1": 3.2245495,
        "q3": 3.691943,
        "min": 3.04146,
        "max": 6.294939,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.091486,
        "q1": 3.0666450000000003,
        "q3": 3.5592125,
        "min": 3.041804,
        "max": 4.026939,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 17.803371,
        "q1": 16.0129455,
        "q3": 18.054535,
        "min": 14.22252,
        "max": 18.305699,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 43.304119957610965,
        "q1": 40.77107098419219,
        "q3": 47.1407370059751,
        "min": 38.23802201077342,
        "max": 50.97735405433923,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.595703125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 290.8054584184545,
        "q1": 262.29768861441335,
        "q3": 292.76497159571284,
        "min": 233.78991881037214,
        "max": 294.7244847729711,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.305699,
        "first_ms": 4.026939,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.386722,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.053144,
          6.294939,
          3.373161,
          5.374983,
          3.257735
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:serial/sparse-uniform/1",
      "language": "rust",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.490188,
        "q1": 3.1828890000000003,
        "q3": 4.33788,
        "min": 1.88797,
        "max": 5.600513,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.125692,
        "q1": 3.8169364999999997,
        "q3": 4.7039825,
        "min": 3.508181,
        "max": 5.282273,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 34.696939,
        "q1": 29.436919500000002,
        "q3": 36.9134905,
        "min": 24.1769,
        "max": 39.130042,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 61.31716107483953,
        "q1": 56.57725752098486,
        "q3": 69.03966155368835,
        "min": 51.837353967130184,
        "max": 76.76216203253716,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.326171875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 280.32804660801384,
        "q1": 247.16655121732836,
        "q3": 289.87483292150114,
        "min": 214.0050558266429,
        "max": 299.4216192349885,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 24.1769,
        "first_ms": 3.508181,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.836246,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.526667,
          3.577382,
          3.441548,
          3.152703,
          3.618486
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:serial/sparse-skew/1",
      "language": "rust",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.627838,
        "q1": 2.5088195,
        "q3": 3.7262934999999997,
        "min": 2.365518,
        "max": 9.253404,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.503782,
        "q1": 3.2736935000000003,
        "q3": 4.2538225,
        "min": 3.043605,
        "max": 5.003863,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 25.668947,
        "q1": 22.033725,
        "q3": 32.486443,
        "min": 18.398503,
        "max": 39.303939,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 63.02131398115307,
        "q1": 51.669378008227795,
        "q3": 68.75321798725054,
        "min": 40.31744203530252,
        "max": 74.485121993348,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.337890625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 338.38668732630185,
        "q1": 245.58648956009495,
        "q3": 342.3684743566903,
        "min": 152.786291793888,
        "max": 346.3502613870787,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "serial",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 39.303939,
        "first_ms": 3.043605,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.775995,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.515872,
          2.533084,
          4.202533,
          2.488315,
          2.501767
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/dense/1",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.986453,
        "q1": 3.727973,
        "q3": 5.453571,
        "min": 3.220568,
        "max": 7.665003,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.459189,
        "q1": 3.908499,
        "q3": 4.9582809999999995,
        "min": 3.357809,
        "max": 5.457373,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 25.834635,
        "q1": 22.335589,
        "q3": 28.6442325,
        "min": 18.836543,
        "max": 31.45383,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 65.39261003490537,
        "q1": 55.735897505655885,
        "q3": 67.49442248838022,
        "min": 46.079184976406395,
        "max": 69.59623494185507,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.6552734375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 239.5479136790051,
        "q1": 201.53414244166467,
        "q3": 253.9628139148972,
        "min": 163.52037120432425,
        "max": 268.3777141507893,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 31.45383,
        "first_ms": 4.459189,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 20.872651,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.261999,
          4.738065,
          3.236642,
          3.94299,
          5.659237
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-uniform/1",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.81718,
        "q1": 2.364483,
        "q3": 3.3158174999999996,
        "min": 2.190545,
        "max": 4.073078,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.022215,
        "q1": 2.8723695,
        "q3": 3.4834899999999998,
        "min": 2.722524,
        "max": 3.944765,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 17.675537,
        "q1": 17.563195,
        "q3": 20.672973499999998,
        "min": 17.450853,
        "max": 23.67041,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 47.81755502335727,
        "q1": 43.19151054369286,
        "q3": 48.29210753086954,
        "min": 38.56546606402844,
        "max": 48.766660038381815,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.3740234375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 368.96728344959615,
        "q1": 323.8743837485532,
        "q3": 372.58874763474444,
        "min": 278.7814840475102,
        "max": 376.2102118198928,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.675537,
        "first_ms": 3.944765,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.935194,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.468171,
          3.163464,
          4.073078,
          3.078034,
          3.628041
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-skew/1",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.39602,
        "q1": 2.3776314999999997,
        "q3": 3.772188,
        "min": 2.290632,
        "max": 4.002592,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.089333,
        "q1": 2.84271,
        "q3": 3.4696545,
        "min": 2.596087,
        "max": 3.849976,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 17.589034,
        "q1": 17.5677275,
        "q3": 18.094767,
        "min": 17.546421,
        "max": 18.6005,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 42.81719995196909,
        "q1": 41.23846552101895,
        "q3": 45.302411483135074,
        "min": 39.65973109006882,
        "max": 47.78762301430106,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.4052734375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 324.93554740949355,
        "q1": 291.346534670525,
        "q3": 338.7229822224137,
        "min": 257.7575219315565,
        "max": 352.51041703533383,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.589034,
        "first_ms": 2.596087,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.387667,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.386784,
          4.002592,
          2.335253,
          2.405621,
          3.779018
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/dense/2",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.293505,
        "q1": 3.1902825,
        "q3": 6.9183335,
        "min": 3.026471,
        "max": 8.175983,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.010774,
        "q1": 3.7030069999999995,
        "q3": 6.320072,
        "min": 3.39524,
        "max": 8.62937,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 18.986252,
        "q1": 18.4032825,
        "q3": 25.080315499999998,
        "min": 17.820313,
        "max": 31.174379,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 46.924202935770154,
        "q1": 46.858954010531306,
        "q3": 68.43393348390236,
        "min": 46.79370508529246,
        "max": 89.94366403203458,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.6865234375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 268.18283654518933,
        "q1": 200.7332461971573,
        "q3": 273.19605845841176,
        "min": 133.28365584912527,
        "max": 278.2092803716342,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 17.820313,
        "first_ms": 4.010774,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 17.972082,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.102214,
          4.352911,
          3.026471,
          3.166826,
          4.293505
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-uniform/2",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.616247,
        "q1": 3.1823145,
        "q3": 3.9365975000000004,
        "min": 2.326351,
        "max": 4.935107,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.919035,
        "q1": 3.2437145000000003,
        "q3": 4.0436295,
        "min": 2.568394,
        "max": 4.168224,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.446315,
        "q1": 18.6350365,
        "q3": 19.493806,
        "min": 17.823758,
        "max": 19.541297,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 47.938960953615606,
        "q1": 46.375541016459465,
        "q3": 48.73321042396128,
        "min": 44.812121079303324,
        "max": 49.52745989430696,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.4365234375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 265.40293819099975,
        "q1": 257.7819771645185,
        "q3": 291.1335642439859,
        "min": 250.16101613803724,
        "max": 316.86419029697214,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.823758,
        "first_ms": 2.568394,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 15.779631,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.395305,
          4.935107,
          2.326351,
          2.991595,
          2.61867
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-skew/2",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.13927,
        "q1": 2.3888425,
        "q3": 5.090405,
        "min": 2.274835,
        "max": 7.396762,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.880559,
        "q1": 2.832114,
        "q3": 6.112815,
        "min": 2.783669,
        "max": 9.345071,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.32193,
        "q1": 19.208959999999998,
        "q3": 22.743402,
        "min": 19.09599,
        "max": 26.164874,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 42.02974995132536,
        "q1": 41.733450489118695,
        "q3": 58.696638443507254,
        "min": 41.437151026912034,
        "max": 75.36352693568915,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.4169921875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 327.6444643419287,
        "q1": 245.09006233290188,
        "q3": 334.15506320413454,
        "min": 162.53566032387505,
        "max": 340.6656620663403,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.32193,
        "first_ms": 2.783669,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.677147,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.557246,
          2.431051,
          2.381459,
          3.13927,
          2.679189
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/dense/4",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.269141,
        "q1": 3.9628845000000004,
        "q3": 4.476896999999999,
        "min": 3.004103,
        "max": 5.213468,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.457235,
        "q1": 4.1807935,
        "q3": 4.4819505,
        "min": 3.904352,
        "max": 4.506666,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.61854,
        "q1": 19.47861,
        "q3": 21.177213000000002,
        "min": 18.33868,
        "max": 21.735886,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 51.092911045998335,
        "q1": 49.88414898980409,
        "q3": 52.4697620421648,
        "min": 48.67538693360984,
        "max": 53.84661303833127,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.6630859375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 237.77137084325707,
        "q1": 234.91896970846614,
        "q3": 243.55713252350859,
        "min": 232.06656857367523,
        "max": 249.3428942037601,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 18.33868,
        "first_ms": 3.904352,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 21.028604,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.878546,
          5.213468,
          3.286553,
          4.395623,
          3.219862
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-uniform/4",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.595851,
        "q1": 2.5389595,
        "q3": 3.0911105,
        "min": 2.330534,
        "max": 4.10678,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.518399,
        "q1": 3.0505430000000002,
        "q3": 3.8261760000000002,
        "min": 2.582687,
        "max": 4.133953,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 17.806039,
        "q1": 17.4345155,
        "q3": 17.916344,
        "min": 17.062992,
        "max": 18.026649,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 41.84831399470568,
        "q1": 39.98418099945411,
        "q3": 42.92716196505353,
        "min": 38.120048004202545,
        "max": 44.00600993540138,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.3857421875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 334.99066615506894,
        "q1": 324.8077820296448,
        "q3": 348.5431203513089,
        "min": 314.62489790422063,
        "max": 362.0955745475489,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.062992,
        "first_ms": 2.582687,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.808509,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.902593,
          2.561549,
          2.561896,
          2.414389,
          2.877931
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:std-scoped/sparse-skew/4",
      "language": "rust",
      "backend": "std-scoped",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.803371,
        "q1": 3.2734674999999998,
        "q3": 4.1710895,
        "min": 2.50392,
        "max": 5.844248,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.91337,
        "q1": 2.870333,
        "q3": 3.5529529999999996,
        "min": 2.827296,
        "max": 4.192536,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 19.859185,
        "q1": 19.3075805,
        "q3": 20.563218,
        "min": 18.755976,
        "max": 21.267251,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 49.205602961592376,
        "q1": 47.14806849369779,
        "q3": 51.59484245814383,
        "min": 45.09053402580321,
        "max": 53.984081954695284,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.4326171875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 252.85765818105025,
        "q1": 243.10916524750704,
        "q3": 268.2063402509673,
        "min": 233.36067231396382,
        "max": 283.5550223208843,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "std-scoped",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.859185,
        "first_ms": 2.827296,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.773971,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.980064,
          3.209934,
          2.50392,
          5.844248,
          3.735216
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/dense/1",
      "language": "rust",
      "backend": "rayon",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.241493,
        "q1": 3.3415665,
        "q3": 4.7957160000000005,
        "min": 3.25293,
        "max": 9.727015,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.922275,
        "q1": 4.1986235,
        "q3": 6.061866,
        "min": 3.474972,
        "max": 7.201457,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 28.804273,
        "q1": 23.9432415,
        "q3": 31.532911,
        "min": 19.08221,
        "max": 34.261549,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 58.368397993035614,
        "q1": 54.28018094971776,
        "q3": 67.88452650653198,
        "min": 50.191963906399906,
        "max": 77.40065502002835,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.669921875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 255.1073772461567,
        "q1": 209.58164545694882,
        "q3": 257.9877135802203,
        "min": 164.05591366774095,
        "max": 260.868049914284,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 28.804273,
        "first_ms": 3.474972,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.166778,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.381797,
          3.622397,
          3.358109,
          4.470873,
          3.296307
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-uniform/1",
      "language": "rust",
      "backend": "rayon",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.515221,
        "q1": 3.2321475,
        "q3": 5.8952415,
        "min": 2.345968,
        "max": 9.704766,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.903259,
        "q1": 3.5237619999999996,
        "q3": 4.778354,
        "min": 3.144265,
        "max": 5.653449,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 26.124048,
        "q1": 22.101359,
        "q3": 29.652232499999997,
        "min": 18.07867,
        "max": 33.180417,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 54.76689396891743,
        "q1": 47.56548849400133,
        "q3": 66.57888443442062,
        "min": 40.36408301908523,
        "max": 78.3908748999238,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.376953125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 277.2199035241469,
        "q1": 211.81075221575446,
        "q3": 308.8878261820048,
        "min": 146.401600907362,
        "max": 340.55574883986276,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 26.124048,
        "first_ms": 3.903259,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.036223,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.559489,
          3.507681,
          3.170463,
          3.515221,
          3.713513
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-skew/1",
      "language": "rust",
      "backend": "rayon",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.870479,
        "q1": 3.0404394999999997,
        "q3": 8.353569,
        "min": 2.3611,
        "max": 11.909592,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.276617,
        "q1": 3.8140039999999997,
        "q3": 6.7593285,
        "min": 3.351391,
        "max": 9.24204,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 40.936575,
        "q1": 29.376303,
        "q3": 44.30466,
        "min": 17.816031,
        "max": 47.672745,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 71.27843191847205,
        "q1": 55.40692340582609,
        "q3": 93.85077649494633,
        "min": 39.53541489318013,
        "max": 116.42312107142061,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.349609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 256.76266646180426,
        "q1": 176.54372788674118,
        "q3": 310.4417812321442,
        "min": 96.32478931167807,
        "max": 364.1208960024842,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 17.816031,
        "first_ms": 3.351391,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.731703,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.393457,
          2.811791,
          3.269088,
          2.400829,
          2.3611
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/dense/2",
      "language": "rust",
      "backend": "rayon",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 7.091351,
        "q1": 6.557506500000001,
        "q3": 8.496619500000001,
        "min": 3.212008,
        "max": 10.872655,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 8.470848,
        "q1": 6.1010865,
        "q3": 11.1419155,
        "min": 3.731325,
        "max": 13.812983,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 30.395792,
        "q1": 28.797728499999998,
        "q3": 31.2045955,
        "min": 27.199665,
        "max": 32.013399,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 91.42737893853337,
        "q1": 78.85417644865811,
        "q3": 91.78948844783008,
        "min": 66.28097395878285,
        "max": 92.1515979571268,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.681640625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 125.071350078436,
        "q1": 120.98956952993845,
        "q3": 153.66916310749318,
        "min": 116.9077889814409,
        "max": 182.26697613655034,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 32.013399,
        "first_ms": 8.470848,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 42.76875,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          6.815882,
          7.119204,
          10.872655,
          10.713562,
          7.211421
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-uniform/2",
      "language": "rust",
      "backend": "rayon",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 5.870027,
        "q1": 3.9426755,
        "q3": 6.669525500000001,
        "min": 3.255342,
        "max": 10.856534,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 5.397751,
        "q1": 4.6039025,
        "q3": 6.9387810000000005,
        "min": 3.810054,
        "max": 8.479811,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 29.194561,
        "q1": 23.994593000000002,
        "q3": 32.505184,
        "min": 18.794625,
        "max": 35.815807,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 82.65505696181208,
        "q1": 64.4885859801434,
        "q3": 83.88052345253527,
        "min": 46.32211499847472,
        "max": 85.10598994325846,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.423828125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 139.38427997886154,
        "q1": 135.7065072506985,
        "q3": 202.2754899708134,
        "min": 132.0287345225355,
        "max": 265.1666999627653,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.794625,
        "first_ms": 3.810054,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 18.856063,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          4.135203,
          3.255342,
          3.750148,
          3.603944,
          3.576185
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-skew/2",
      "language": "rust",
      "backend": "rayon",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.298884,
        "q1": 2.610539,
        "q3": 3.3661425,
        "min": 2.073105,
        "max": 4.043313,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.886385,
        "q1": 2.87512,
        "q3": 3.1780885000000003,
        "min": 2.863855,
        "max": 3.469792,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 18.606489,
        "q1": 18.137652,
        "q3": 18.713684999999998,
        "min": 17.668815,
        "max": 18.820881,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 43.12821105122566,
        "q1": 42.35397151205689,
        "q3": 44.135445554275066,
        "min": 41.57973197288811,
        "max": 45.14268005732447,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.408203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 306.5072657240833,
        "q1": 288.19778190774633,
        "q3": 338.4195196470264,
        "min": 269.88829809140935,
        "max": 370.3317735699694,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 18.606489,
        "first_ms": 2.886385,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 13.501407,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.451363,
          3.388572,
          2.72677,
          2.349196,
          2.073105
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/dense/4",
      "language": "rust",
      "backend": "rayon",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.70269,
        "q1": 3.958264,
        "q3": 6.5518445,
        "min": 3.295672,
        "max": 14.964834,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.747382,
        "q1": 4.030941,
        "q3": 4.949953,
        "min": 3.3145,
        "max": 5.152524,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 20.343838,
        "q1": 20.0530075,
        "q3": 21.0420965,
        "min": 19.762177,
        "max": 21.740355,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 59.40498295240104,
        "q1": 54.40144147723913,
        "q3": 68.06136347586289,
        "min": 49.39790000207722,
        "max": 76.71774399932474,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 32.677734375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 213.00125585540454,
        "q1": 158.4534266073032,
        "q3": 238.02719311092636,
        "min": 103.90559735920182,
        "max": 263.0531303664482,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 20.343838,
        "first_ms": 4.747382,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 19.007567,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          3.97016,
          3.934241,
          3.295672,
          3.827737,
          3.946368
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "logical_data_bytes": 33587200,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-uniform/4",
      "language": "rust",
      "backend": "rayon",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.438846,
        "q1": 2.8819955000000004,
        "q3": 10.484466000000001,
        "min": 2.164096,
        "max": 19.209879,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.570444,
        "q1": 4.048687,
        "q3": 10.3372365,
        "min": 3.52693,
        "max": 16.104029,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 18.906504,
        "q1": 18.781793,
        "q3": 19.366732,
        "min": 18.657082,
        "max": 19.82696,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 74.21900099143386,
        "q1": 59.7932210075669,
        "q3": 82.24422147031873,
        "min": 45.36744102369994,
        "max": 90.26944194920361,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.412109375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 150.3849448548934,
        "q1": 118.77805844401912,
        "q3": 247.5996902763108,
        "min": 87.17117203314484,
        "max": 344.81443569772824,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 19.82696,
        "first_ms": 3.52693,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 14.500553,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.812838,
          2.601779,
          3.438846,
          2.164096,
          2.951153
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "rust:rayon/sparse-skew/4",
      "language": "rust",
      "backend": "rayon",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.875604,
        "q1": 2.2973465,
        "q3": 6.904329499999999,
        "min": 2.008408,
        "max": 10.730877,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 6.399296,
        "q1": 4.501926,
        "q3": 7.3824745,
        "min": 2.604556,
        "max": 8.365653,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 37.1812,
        "q1": 28.001701999999998,
        "q3": 38.791837,
        "min": 18.822204,
        "max": 40.402474,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 76.99722703546286,
        "q1": 60.029300511814654,
        "q3": 86.21900650905445,
        "min": 43.06137398816645,
        "max": 95.44078598264605,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 25.439453125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 246.98267441237266,
        "q1": 188.89181915018582,
        "q3": 289.1134692519423,
        "min": 130.80096388799902,
        "max": 331.2442640915119,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "language": "rust",
        "backend": "rayon",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 40.402474,
        "first_ms": 6.399296,
        "stress_seconds_requested": 0,
        "elapsed_warm_ms": 20.244335,
        "iterations": 5,
        "validations": "all",
        "samples_ms": [
          2.219803,
          4.781458,
          8.702245,
          2.008408,
          2.120285
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "logical_data_bytes": 25952264,
        "rayon_version": "1.12.0"
      }
    },
    {
      "id": "python:serial/dense/1",
      "language": "python",
      "backend": "serial",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 384.1947439359501,
        "q1": 345.03265493549407,
        "q3": 412.13260195218027,
        "min": 248.50433401297778,
        "max": 437.03902303241193,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 368.94300393760204,
        "q1": 348.10607944382355,
        "q3": 403.51607400225475,
        "min": 327.26915495004505,
        "max": 438.08914406690747,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1035.1449369918555,
        "q1": 1012.8552035312168,
        "q3": 1133.673155971337,
        "min": 990.5654700705782,
        "max": 1232.2013749508187,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3375.566439004615,
        "q1": 3217.7835634793155,
        "q3": 3563.2771115051582,
        "min": 3060.000687954016,
        "max": 3750.9877840057015,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 43.248046875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 2.5912783969263633,
        "q1": 2.546461646768711,
        "q3": 2.7584622519122517,
        "min": 2.5016448966110585,
        "max": 2.92564610689814,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 990.5654700705782,
        "first_ms": 327.26915495004505,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1709.0242009144276,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          328.0389659339562,
          345.1197649119422,
          351.35353601071984,
          344.94554495904595,
          335.13414103072137
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      }
    },
    {
      "id": "python:serial/sparse-uniform/1",
      "language": "python",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 209.25709698349237,
        "q1": 202.39385101012886,
        "q3": 220.40895401732996,
        "min": 200.41312999092042,
        "max": 223.75824104528874,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 209.40250402782112,
        "q1": 204.05954000307247,
        "q3": 214.98958754818887,
        "min": 198.71657597832382,
        "max": 220.5766710685566,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 813.3335050661117,
        "q1": 793.7474440550432,
        "q3": 827.9472340364009,
        "min": 774.1613830439746,
        "max": 842.5609630066901,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2182.468288927339,
        "q1": 2137.6254999195226,
        "q3": 2238.328813458793,
        "min": 2092.782710911706,
        "max": 2294.1893379902467,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.2236328125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.486950619989441,
        "q1": 4.3703213197852655,
        "q3": 4.572742364103162,
        "min": 4.25369201958109,
        "max": 4.658534108216882,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 813.3335050661117,
        "first_ms": 209.40250402782112,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1114.3425509799272,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          209.25709698349237,
          206.23234403319657,
          210.3280610172078,
          209.10032698884606,
          210.15124500263482
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:serial/sparse-skew/1",
      "language": "python",
      "backend": "serial",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 208.2193500827998,
        "q1": 204.67620505951345,
        "q3": 263.5531095438637,
        "min": 200.25739108677953,
        "max": 271.462781005539,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 209.92281893268228,
        "q1": 204.56110395025462,
        "q3": 240.0596379302442,
        "min": 199.19938896782696,
        "max": 270.19645692780614,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 841.3866889895871,
        "q1": 823.2183760264888,
        "q3": 895.9808709914796,
        "min": 805.0500630633906,
        "max": 950.5750529933721,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2226.8469349946827,
        "q1": 2179.1956769884564,
        "q3": 2471.0490769939497,
        "min": 2131.54441898223,
        "max": 2715.2512189932168,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.224609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.4758988495407355,
        "q1": 4.000551749903323,
        "q3": 4.552717764306033,
        "min": 3.52520465026591,
        "max": 4.6295366790713315,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 841.3866889895871,
        "first_ms": 209.92281893268228,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1117.0940559823066,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          208.2193500827998,
          207.231329055503,
          212.45986700523645,
          209.55645700450987,
          207.4480860028416
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/dense/1",
      "language": "python",
      "backend": "thread",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 436.5549370413646,
        "q1": 351.9263949710876,
        "q3": 450.7292460766621,
        "min": 351.00913094356656,
        "max": 468.55706500355154,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 371.98938406072557,
        "q1": 362.6630750368349,
        "q3": 406.28274250775576,
        "min": 353.3367660129443,
        "max": 440.57610095478594,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1061.6817019181326,
        "q1": 1057.4374129646458,
        "q3": 1174.8242769390345,
        "min": 1053.193124011159,
        "max": 1287.9668519599363,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3776.0797599330544,
        "q1": 3494.01375092566,
        "q3": 3886.2818994675763,
        "min": 3211.947741918266,
        "max": 3996.484039002098,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 43.28125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 2.279059113292171,
        "q1": 2.234835432662674,
        "q3": 2.5582574015151813,
        "min": 2.1906117520331776,
        "max": 2.8374556897381913,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1287.9668519599363,
        "first_ms": 440.57610095478594,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2193.8878069631755,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          449.35067906044424,
          433.189157047309,
          433.73095500282943,
          436.5549370413646,
          438.3698560995981
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-uniform/1",
      "language": "python",
      "backend": "thread",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 205.92900994233787,
        "q1": 176.72996746841818,
        "q3": 207.68255548318848,
        "min": 146.09801000915468,
        "max": 227.60133200790733,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 205.5757089983672,
        "q1": 175.83341849967837,
        "q3": 207.03769201645628,
        "min": 146.09112800098956,
        "max": 208.49967503454536,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 816.3534270133823,
        "q1": 811.5699319751002,
        "q3": 825.364880496636,
        "min": 806.7864369368181,
        "max": 834.3763339798898,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2172.180718043819,
        "q1": 2023.8302189973183,
        "q3": 2183.2347700255923,
        "min": 1875.4797199508175,
        "max": 2194.2888220073655,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.2197265625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.539678654341669,
        "q1": 4.517953330244833,
        "q3": 5.183637754445771,
        "min": 4.496228006147995,
        "max": 5.827596854549873,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 834.3763339798898,
        "first_ms": 208.49967503454536,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1101.3995440443978,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          207.48523296788335,
          205.1974789937958,
          206.1271530110389,
          205.15290088951588,
          207.8798779984936
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-skew/1",
      "language": "python",
      "backend": "thread",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 206.87277999240905,
        "q1": 202.07058847881854,
        "q3": 214.15137650910765,
        "min": 196.0096909897402,
        "max": 285.46207305043936,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 204.32733290363103,
        "q1": 203.39032995980233,
        "q3": 207.43272342951968,
        "min": 202.45332701597363,
        "max": 210.53811395540833,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 823.873660992831,
        "q1": 808.110810990911,
        "q3": 825.9942589793354,
        "min": 792.347960988991,
        "max": 828.1148569658399,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2206.2825600150973,
        "q1": 2154.797401977703,
        "q3": 2272.8115979698487,
        "min": 2103.3122439403087,
        "max": 2339.3406359246,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.2236328125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.423627909479643,
        "q1": 4.20781483117928,
        "q3": 4.57001663742542,
        "min": 3.992001752878917,
        "max": 4.7164053653711955,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 828.1148569658399,
        "first_ms": 210.53811395540833,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1252.5044600479305,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          261.38196501415223,
          285.46207305043936,
          207.47873303480446,
          209.01513996068388,
          206.87277999240905
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/dense/2",
      "language": "python",
      "backend": "thread",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 347.89381292648613,
        "q1": 274.4161324808374,
        "q3": 372.57901253178716,
        "min": 229.1884000878781,
        "max": 382.475706981495,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 373.02993703633547,
        "q1": 352.1687189931981,
        "q3": 379.0412080124952,
        "min": 331.3075009500608,
        "max": 385.0524789886549,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1033.005633042194,
        "q1": 1014.6446475409903,
        "q3": 1180.8081080089323,
        "min": 996.2836620397866,
        "max": 1328.6105829756707,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3093.079084996134,
        "q1": 3060.520072001964,
        "q3": 3215.085409465246,
        "min": 3027.9610590077937,
        "max": 3337.0917339343578,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 43.294921875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 2.889808719892745,
        "q1": 2.767912131350732,
        "q3": 3.4296762863347388,
        "min": 2.6460155428087195,
        "max": 3.969543852776732,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1328.6105829756707,
        "first_ms": 385.0524789886549,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1259.5905689522624,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          302.70623101387173,
          229.1884000878781,
          246.12603394780308,
          238.36051707621664,
          240.27377599850297
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-uniform/2",
      "language": "python",
      "backend": "thread",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 210.16937296371907,
        "q1": 203.15374998608604,
        "q3": 215.39216849487275,
        "min": 150.44158906675875,
        "max": 222.36070199869573,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 220.10205092374235,
        "q1": 210.98676597466692,
        "q3": 220.5225529614836,
        "min": 201.8714810255915,
        "max": 220.94305499922484,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 808.0680640414357,
        "q1": 804.4099605176598,
        "q3": 818.2669859961607,
        "min": 800.7518569938838,
        "max": 828.4659079508856,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2144.1912170266733,
        "q1": 2111.21176648885,
        "q3": 2196.1797475232743,
        "min": 2078.232315951027,
        "max": 2248.1682780198753,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.2255859375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.563846565113287,
        "q1": 4.471824806700531,
        "q3": 4.834570855122562,
        "min": 4.3798030482877754,
        "max": 5.105295145131837,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 800.7518569938838,
        "first_ms": 201.8714810255915,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1095.567067968659,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          203.30503396689892,
          207.135119009763,
          203.00246600527316,
          210.16937296371907,
          205.22145798895508
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-skew/2",
      "language": "python",
      "backend": "thread",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 207.15396897867322,
        "q1": 204.69316747039557,
        "q3": 217.75681053986773,
        "min": 201.75628294236958,
        "max": 227.9725450789556,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 211.49809891358018,
        "q1": 207.1763714775443,
        "q3": 215.57318943087012,
        "min": 202.85464404150844,
        "max": 219.64827994816005,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 820.4807809088379,
        "q1": 817.2026704414748,
        "q3": 844.0016089589335,
        "min": 813.9245599741116,
        "max": 867.5224370090291,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2199.6769360266626,
        "q1": 2183.7718650349416,
        "q3": 2258.7648634798825,
        "min": 2167.8667940432206,
        "max": 2317.8527909331024,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.2314453125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.487641121802564,
        "q1": 4.384483099037961,
        "q3": 4.537997348124766,
        "min": 4.281325076273359,
        "max": 4.588353574446969,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 820.4807809088379,
        "first_ms": 211.49809891358018,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1089.7154979174957,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          202.08023593295366,
          205.0095689482987,
          207.62760599609464,
          204.37676599249244,
          204.0412169881165
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/dense/4",
      "language": "python",
      "backend": "thread",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 377.76770896743983,
        "q1": 330.0899915047921,
        "q3": 392.19081052578986,
        "min": 245.22225908003747,
        "max": 545.3514020191506,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 367.06798803061247,
        "q1": 333.8205570471473,
        "q3": 373.18824301473796,
        "min": 300.57312606368214,
        "max": 379.30849799886346,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1048.7435009563342,
        "q1": 1034.969940490555,
        "q3": 1053.5485264845192,
        "min": 1021.1963800247759,
        "max": 1058.3535520127043,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3444.3514940794557,
        "q1": 3154.187379986979,
        "q3": 3462.203274539206,
        "min": 2864.0232658945024,
        "max": 3480.0550549989566,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 43.3408203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 2.5486599394285414,
        "q1": 2.4957320331548782,
        "q3": 2.9852742524462537,
        "min": 2.442804126881215,
        "max": 3.421888565463966,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1058.3535520127043,
        "first_ms": 379.30849799886346,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1961.8152750190347,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          377.76770896743983,
          396.5117339976132,
          401.1584360850975,
          396.741695003584,
          385.3122319560498
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-uniform/4",
      "language": "python",
      "backend": "thread",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 212.9160879412666,
        "q1": 176.10182898351923,
        "q3": 232.99590405076742,
        "min": 139.06475494150072,
        "max": 240.06267404183745,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 229.71244505606592,
        "q1": 221.19344549719244,
        "q3": 234.5850525307469,
        "min": 212.67444593831897,
        "max": 239.4576600054279,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1014.5716250408441,
        "q1": 922.6410810369998,
        "q3": 1015.6369549804367,
        "min": 830.7105370331556,
        "max": 1016.7022849200293,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2215.2193599613383,
        "q1": 2182.140819961205,
        "q3": 2391.964697977528,
        "min": 2149.062279961072,
        "max": 2568.7100359937176,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.220703125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.450292908609319,
        "q1": 4.234006467192144,
        "q3": 5.225213712482958,
        "min": 4.0177200257749695,
        "max": 6.000134516356598,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 830.7105370331556,
        "first_ms": 212.67444593831897,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1123.5215530032292,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          209.99859203584492,
          213.16046407446265,
          210.51585406530648,
          209.6565399551764,
          212.9160879412666
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:thread/sparse-skew/4",
      "language": "python",
      "backend": "thread",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 222.56641602143645,
        "q1": 204.9397504888475,
        "q3": 269.7908920235932,
        "min": 200.64173196442425,
        "max": 274.2250619921833,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 221.33027494419366,
        "q1": 213.8940294389613,
        "q3": 252.22800549818203,
        "min": 206.45778393372893,
        "max": 283.1257360521704,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 836.4941939944401,
        "q1": 817.8556819912046,
        "q3": 925.5219275364652,
        "min": 799.2171699879691,
        "max": 1014.5496610784903,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2327.374243061058,
        "q1": 2233.4564800257795,
        "q3": 2565.71661203634,
        "min": 2139.5387169905007,
        "max": 2804.0589810116217,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 36.224609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 4.226980153676416,
        "q1": 3.8617364066279247,
        "q3": 4.419671617806507,
        "min": 3.4964926595794332,
        "max": 4.6123630819366,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 836.4941939944401,
        "first_ms": 221.33027494419366,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1182.877567014657,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          226.42951598390937,
          222.56641602143645,
          224.52221508137882,
          222.04406303353608,
          222.10642998106778
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        }
      }
    },
    {
      "id": "python:process-copy/dense/1",
      "language": "python",
      "backend": "process-copy",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 488.23892208747566,
        "q1": 472.96458948403597,
        "q3": 537.3361354577355,
        "min": 465.2786289807409,
        "max": 547.1321659861133,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 562.2132769785821,
        "q1": 561.9175725150853,
        "q3": 613.249066984281,
        "min": 561.6218680515885,
        "max": 664.2848569899797,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1022.1713660284877,
        "q1": 1018.4125060332008,
        "q3": 1039.2395629896782,
        "min": 1014.6536460379139,
        "max": 1056.3077599508688,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 4103.6585019901395,
        "q1": 4050.869997998234,
        "q3": 4304.907445504796,
        "min": 3998.0814940063283,
        "max": 4506.156389019452,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 187.9755859375,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 2.030965849481358,
        "q1": 1.9379718751761432,
        "q3": 2.074937684946117,
        "min": 1.8449779008709288,
        "max": 2.1189095204108765,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1022.1713660284877,
        "first_ms": 562.2132769785821,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2461.8828530656174,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          485.81622599158436,
          520.7985689630732,
          484.05319498851895,
          478.0322159640491,
          488.7827590573579
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-uniform/1",
      "language": "python",
      "backend": "process-copy",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 285.677285050042,
        "q1": 237.81022045295686,
        "q3": 382.202188950032,
        "min": 204.41467594355345,
        "max": 418.8112650299445,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 412.1586650144309,
        "q1": 366.50588450720534,
        "q3": 739.492651540786,
        "min": 320.8531039999798,
        "max": 1066.8266380671412,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 824.247932061553,
        "q1": 813.2660735282116,
        "q3": 928.8747520186007,
        "min": 802.2842149948701,
        "max": 1033.5015719756484,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2785.0264291046187,
        "q1": 2608.4513250389136,
        "q3": 3544.156746531371,
        "min": 2431.8762209732085,
        "max": 4303.287063958123,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 200.5419921875,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 3.393357808467765,
        "q1": 2.910531325921038,
        "q3": 3.8016393466760157,
        "min": 2.4277048433743107,
        "max": 4.209920884884267,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 824.247932061553,
        "first_ms": 412.1586650144309,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1473.46677898895,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          289.4783599767834,
          285.677285050042,
          266.7606499744579,
          278.7270799744874,
          266.9619229855016
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-skew/1",
      "language": "python",
      "backend": "process-copy",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 273.14355003181845,
        "q1": 239.57644251640886,
        "q3": 289.12030602805316,
        "min": 201.1630500201136,
        "max": 834.9793370580301,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 501.9086350221187,
        "q1": 407.2863819892518,
        "q3": 820.8223115070723,
        "min": 312.66412895638496,
        "max": 1139.735987992026,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 807.9197860788554,
        "q1": 655.3253644960932,
        "q3": 904.06485152198,
        "min": 502.7309429133311,
        "max": 1000.2099169651046,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2884.4877049559727,
        "q1": 2554.984439513646,
        "q3": 3544.235230947379,
        "min": 2225.4811740713194,
        "max": 4203.982756938785,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 200.80859375,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 3.3270388442186256,
        "q1": 2.9217346369934405,
        "q3": 3.5662570939165326,
        "min": 2.5164304297682554,
        "max": 3.8054753436144395,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 807.9197860788554,
        "first_ms": 501.9086350221187,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1502.837879001163,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          297.59279591962695,
          279.80662893969566,
          282.4394170893356,
          295.8011949667707,
          273.14355003181845
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/dense/2",
      "language": "python",
      "backend": "process-copy",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 350.69663904141635,
        "q1": 333.1102689844556,
        "q3": 384.05175000661984,
        "min": 304.15881401859224,
        "max": 442.97629594802856,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 518.0587419308722,
        "q1": 479.44753599585965,
        "q3": 623.3200834831223,
        "min": 440.8363300608471,
        "max": 728.5814250353724,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1024.4837930658832,
        "q1": 867.2117305104621,
        "q3": 1033.801208541263,
        "min": 709.9396679550409,
        "max": 1043.1186240166426,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3233.285915106535,
        "q1": 3213.889935053885,
        "q3": 3469.169119547587,
        "min": 3194.493955001235,
        "max": 3705.052323988639,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 233.5537109375,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 2.710841832290505,
        "q1": 2.639254918227917,
        "q3": 2.8743311375437743,
        "min": 2.5676680041653297,
        "max": 3.0378204427970443,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1024.4837930658832,
        "first_ms": 440.8363300608471,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1645.9168980363756,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          354.1240270715207,
          339.2205099808052,
          304.15881401859224,
          314.68542595393956,
          329.7106600366533
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-uniform/2",
      "language": "python",
      "backend": "process-copy",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 299.8073570197448,
        "q1": 280.53262655157596,
        "q3": 326.4685574686155,
        "min": 271.6808350523934,
        "max": 361.4604000467807,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 420.491985976696,
        "q1": 417.314036982134,
        "q3": 438.66045598406345,
        "min": 414.13608798757195,
        "max": 456.8289259914309,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 851.1490810196847,
        "q1": 829.3114110128954,
        "q3": 853.8491755025461,
        "min": 807.4737410061061,
        "max": 856.5492699854076,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2912.013405933976,
        "q1": 2883.661834988743,
        "q3": 3014.941369998269,
        "min": 2855.31026404351,
        "max": 3117.8693340625614,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 171.0791015625,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 3.207640439450424,
        "q1": 3.031864617480694,
        "q3": 3.2573725506475366,
        "min": 2.856088795510963,
        "max": 3.307104661844649,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 851.1490810196847,
        "first_ms": 420.491985976696,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1750.6458510179073,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          330.2291949512437,
          346.9733780948445,
          361.4604000467807,
          312.7934290096164,
          322.70791998598725
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-skew/2",
      "language": "python",
      "backend": "process-copy",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 267.26148603484035,
        "q1": 248.8558935583569,
        "q3": 305.2877960144542,
        "min": 202.07455498166382,
        "max": 370.43733603786677,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 411.34415194392204,
        "q1": 390.2373669552617,
        "q3": 490.75073248241097,
        "min": 369.1305819666013,
        "max": 570.1573130208999,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 781.6206680145115,
        "q1": 763.112235465087,
        "q3": 821.7268764856271,
        "min": 744.6038029156625,
        "max": 861.8330849567428,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2688.004603027366,
        "q1": 2645.7492950139567,
        "q3": 2897.456324484665,
        "min": 2603.4939870005473,
        "max": 3106.908045941964,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 215.216796875,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 3.4469058240884687,
        "q1": 3.315596576154028,
        "q3": 3.6185786317341417,
        "min": 3.1842873282195874,
        "max": 3.790251439379815,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 861.8330849567428,
        "first_ms": 570.1573130208999,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1570.2100610360503,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          320.08097402285784,
          273.0738699901849,
          274.3758560391143,
          290.4946180060506,
          327.9139829101041
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/dense/4",
      "language": "python",
      "backend": "process-copy",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 329.15910496376455,
        "q1": 272.5476610357873,
        "q3": 370.8109574508853,
        "min": 217.80251897871494,
        "max": 397.1546719549224,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 451.9929299131036,
        "q1": 408.50677742855623,
        "q3": 508.9116399176419,
        "min": 365.0206249440089,
        "max": 565.8303499221802,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1029.708594083786,
        "q1": 1010.1003770250827,
        "q3": 1060.5138725368306,
        "min": 990.4921599663794,
        "max": 1091.3191509898752,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3405.059692915529,
        "q1": 3047.3141614929773,
        "q3": 3437.2169659473,
        "min": 2689.5686300704256,
        "max": 3469.3742389790714,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 236.0244140625,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 2.908472871093917,
        "q1": 2.8226070480221406,
        "q3": 3.48328774279468,
        "min": 2.736741224950364,
        "max": 4.058102614495442,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1091.3191509898752,
        "first_ms": 565.8303499221802,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1719.1152269952,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          310.3109629591927,
          363.26718202326447,
          301.1354700429365,
          370.16540800686926,
          371.4565068949014
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-uniform/4",
      "language": "python",
      "backend": "process-copy",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 212.3701290693134,
        "q1": 204.11678397795185,
        "q3": 252.04433349426836,
        "min": 182.01826501172036,
        "max": 297.292627976276,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 424.7305759927258,
        "q1": 402.38517650868744,
        "q3": 434.8280389676802,
        "min": 380.0397770246491,
        "max": 444.9255019426346,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 804.2345619760454,
        "q1": 802.3116599651985,
        "q3": 900.8982355007902,
        "min": 800.3887579543516,
        "max": 997.561909025535,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2603.8185940124094,
        "q1": 2563.697158533614,
        "q3": 2655.781671521254,
        "min": 2523.575723054819,
        "max": 2707.7447490300983,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 172.666015625,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 4.280742188939346,
        "q1": 4.033830058938104,
        "q3": 4.314577283480317,
        "min": 3.7869179289368615,
        "max": 4.348412378021289,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 800.3887579543516,
        "first_ms": 444.9255019426346,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1168.0217540124431,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          212.3701290693134,
          202.59262097533792,
          259.691154002212,
          204.83576005790383,
          227.9663230292499
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-copy/sparse-skew/4",
      "language": "python",
      "backend": "process-copy",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 292.1418610494584,
        "q1": 258.5363640100695,
        "q3": 346.21699049603194,
        "min": 234.20364293269813,
        "max": 755.2039310103282,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 430.2693100180477,
        "q1": 390.4945719987154,
        "q3": 482.6504589873366,
        "min": 350.7198339793831,
        "max": 535.0316079566255,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 822.6592489518225,
        "q1": 809.970174508635,
        "q3": 831.9213014910929,
        "min": 797.2811000654474,
        "max": 841.1833540303633,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3088.6318699922413,
        "q1": 2991.941398533527,
        "q3": 3249.5491034933366,
        "min": 2895.2509270748124,
        "max": 3410.466336994432,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 234.19921875,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 2.7277103422933786,
        "q1": 2.5765216475110035,
        "q3": 3.1595099386702588,
        "min": 2.425332952728629,
        "max": 3.5913095350471393,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-copy",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 841.1833540303633,
        "first_ms": 535.0316079566255,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1392.2498050378636,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          302.8840370243415,
          260.3723850334063,
          241.53898702934384,
          254.55736310686916,
          256.7003429867327
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/dense/1",
      "language": "python",
      "backend": "process-shared",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 460.5883789481595,
        "q1": 313.99929500184953,
        "q3": 469.39645550446585,
        "min": 255.9082160005346,
        "max": 500.1524150138721,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 423.1955960858613,
        "q1": 365.39996054489166,
        "q3": 472.0991380745545,
        "min": 307.604325003922,
        "max": 521.0026800632477,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 656.8041329737753,
        "q1": 656.208124011755,
        "q3": 792.7049854770303,
        "min": 655.6121150497347,
        "max": 928.6058379802853,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 3603.6130680004135,
        "q1": 3064.5351834828034,
        "q3": 3625.4085730179213,
        "min": 2525.4572989651933,
        "max": 3647.204078035429,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 59.8828125,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 2.279298974388672,
        "q1": 2.202532885311518,
        "q3": 2.820843671142482,
        "min": 2.125766796234364,
        "max": 3.3623883678962923,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 655.6121150497347,
        "first_ms": 521.0026800632477,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 2352.0924350013956,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          469.70460494048893,
          463.58202607370913,
          465.322058997117,
          469.08830606844276,
          481.71718802768737
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-uniform/1",
      "language": "python",
      "backend": "process-shared",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 239.22879889141768,
        "q1": 236.58530449029058,
        "q3": 257.2093129856512,
        "min": 235.2206140058115,
        "max": 291.0654579754919,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 303.2615160336718,
        "q1": 297.63317049946636,
        "q3": 315.0111035211012,
        "min": 292.0048249652609,
        "max": 326.76069100853056,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 838.6926769744605,
        "q1": 836.7056684801355,
        "q3": 858.9348425157368,
        "min": 834.7186599858105,
        "max": 879.1770080570132,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2470.461779041216,
        "q1": 2463.3593464968726,
        "q3": 2582.2730535292067,
        "min": 2456.2569139525294,
        "max": 2694.0843280171975,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 52.5986328125,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 3.962762893697756,
        "q1": 3.76267814854517,
        "q3": 3.975951362728344,
        "min": 3.562593403392584,
        "max": 3.989139831758932,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 834.7186599858105,
        "first_ms": 292.0048249652609,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1253.4030419774354,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          235.2206140058115,
          236.43123102374375,
          236.14097794052213,
          235.52718095015734,
          239.22879889141768
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-skew/1",
      "language": "python",
      "backend": "process-shared",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 251.21875992044806,
        "q1": 178.5801264923066,
        "q3": 271.9051939784549,
        "min": 157.71950501948595,
        "max": 320.5612499732524,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 382.30403198394924,
        "q1": 354.3210760108195,
        "q3": 409.5426514977589,
        "min": 326.3381200376898,
        "max": 436.7812710115686,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 958.6752449395135,
        "q1": 909.8015389754437,
        "q3": 962.7992474706843,
        "min": 860.9278330113739,
        "max": 966.9232500018552,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2606.8525910377502,
        "q1": 2586.7975170258433,
        "q3": 2640.730319020804,
        "min": 2566.7424430139363,
        "max": 2674.608047003858,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 52.712890625,
      "max_processes": 3,
      "throughput_calls_per_second": {
        "median": 4.141209698471787,
        "q1": 3.947617391125476,
        "q3": 4.20788500380014,
        "min": 3.7540250837791653,
        "max": 4.274560309128492,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 860.9278330113739,
        "first_ms": 326.3381200376898,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1331.903726910241,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          251.21875992044806,
          251.25836802180856,
          249.6653599664569,
          253.51341103669256,
          256.909942952916
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/dense/2",
      "language": "python",
      "backend": "process-shared",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 259.3750739470124,
        "q1": 253.53596702916548,
        "q3": 264.225596038159,
        "min": 234.77956501301378,
        "max": 403.37925602216274,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 365.05472799763083,
        "q1": 361.90554848872125,
        "q3": 458.6103609763086,
        "min": 358.75636897981167,
        "max": 552.1659939549863,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 1016.5401879930869,
        "q1": 944.6853019762784,
        "q3": 1033.2007720135152,
        "min": 872.83041595947,
        "max": 1049.8613560339436,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2783.741423045285,
        "q1": 2692.099354986567,
        "q3": 3009.684812044725,
        "min": 2600.457286927849,
        "max": 3235.6282010441646,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 69.8916015625,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 3.863502882295755,
        "q1": 3.513822368494445,
        "q3": 3.918619331615459,
        "min": 3.1641418546931344,
        "max": 3.9737357809351637,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 1049.8613560339436,
        "first_ms": 365.05472799763083,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1294.162357924506,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          260.8238860266283,
          266.8942059390247,
          259.3750739470124,
          249.90616191644222,
          254.39303100574762
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-uniform/2",
      "language": "python",
      "backend": "process-shared",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 172.4721910431981,
        "q1": 164.69758842140436,
        "q3": 228.65418496076018,
        "min": 158.52572408039123,
        "max": 230.8874810114503,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 311.4938939688727,
        "q1": 296.35995149146765,
        "q3": 331.00130944512784,
        "min": 281.2260090140626,
        "max": 350.50872492138296,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 814.2680620076135,
        "q1": 810.4944420047104,
        "q3": 830.4524054983631,
        "min": 806.7208220018074,
        "max": 846.6367489891127,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2159.7177360672504,
        "q1": 2145.742272492498,
        "q3": 2307.77986953035,
        "min": 2131.766808917746,
        "max": 2455.8420029934496,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 62.626953125,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 5.454939053015169,
        "q1": 4.7872910663368575,
        "q3": 5.494651144635051,
        "min": 4.119643079658546,
        "max": 5.534363236254933,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 806.7208220018074,
        "first_ms": 350.50872492138296,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1213.6973770102486,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          228.15931495279074,
          230.8874810114503,
          229.27246091421694,
          230.57327000424266,
          229.14905496872962
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-skew/2",
      "language": "python",
      "backend": "process-shared",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 235.07167398929596,
        "q1": 212.54366054199636,
        "q3": 242.85908503225073,
        "min": 151.65546396747231,
        "max": 257.8932299511507,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 378.5831310087815,
        "q1": 319.0791084780358,
        "q3": 388.48392001818866,
        "min": 259.57508594729006,
        "max": 398.3847090275958,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 882.9461509594694,
        "q1": 731.6863635205664,
        "q3": 911.75118094543,
        "min": 580.4265760816634,
        "max": 940.5562109313905,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2639.999584062025,
        "q1": 2306.7013675463386,
        "q3": 2660.593501001131,
        "min": 1973.4031510306522,
        "max": 2681.1874179402366,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 62.669921875,
      "max_processes": 4,
      "throughput_calls_per_second": {
        "median": 4.019327518601959,
        "q1": 3.9639116226578137,
        "q3": 4.398048461202851,
        "min": 3.9084957267136686,
        "max": 4.776769403803743,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 882.9461509594694,
        "first_ms": 378.5831310087815,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1279.2645430890843,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          242.33214696869254,
          239.49277598876506,
          243.38602309580892,
          246.46818300243467,
          236.5935390116647
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/dense/4",
      "language": "python",
      "backend": "process-shared",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 192.3178219003603,
        "q1": 186.20521592674777,
        "q3": 195.6847639521584,
        "min": 174.6403370052576,
        "max": 203.10004393104464,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 323.7717649899423,
        "q1": 320.36156399408355,
        "q3": 348.27580698765814,
        "min": 316.9513629982248,
        "max": 372.779848985374,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 997.6554920431226,
        "q1": 841.138279531151,
        "q3": 1032.8652809839696,
        "min": 684.6210670191795,
        "max": 1068.0750699248165,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2343.6437679920346,
        "q1": 2260.6665134662762,
        "q3": 2390.311490977183,
        "min": 2177.689258940518,
        "max": 2436.9792139623314,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 89.625,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 5.2941175208227245,
        "q1": 5.1757155959719245,
        "q3": 5.315913096489236,
        "min": 5.057313671121124,
        "max": 5.337708672155748,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 997.6554920431226,
        "first_ms": 323.7717649899423,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 944.4444669643417,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          181.39795598108321,
          196.20269897859544,
          195.59253996703774,
          192.39557697437704,
          174.6403370052576
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 33554432
          },
          "x": {
            "format": "d",
            "bytes": 16384
          },
          "output": {
            "format": "d",
            "bytes": 16384
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-uniform/4",
      "language": "python",
      "backend": "process-shared",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 126.84042495675385,
        "q1": 122.3894699360244,
        "q3": 132.5490745366551,
        "min": 116.8947669211775,
        "max": 163.55662094429135,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 379.5663519995287,
        "q1": 341.01099701365456,
        "q3": 384.63604799471796,
        "min": 302.4556420277804,
        "max": 389.7057439899072,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 830.6983310030773,
        "q1": 665.0488215382211,
        "q3": 832.2907644906081,
        "min": 499.39931207336485,
        "max": 833.8831979781389,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2079.263301915489,
        "q1": 1847.130671958439,
        "q3": 2095.6059309537522,
        "min": 1614.998042001389,
        "max": 2111.9485599920154,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 85.1904296875,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 6.5281770734630005,
        "q1": 6.505929035527924,
        "q3": 6.911596027410556,
        "min": 6.483680997592847,
        "max": 7.295014981358111,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 830.6983310030773,
        "first_ms": 379.5663519995287,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 771.1668729316443,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          163.55662094429135,
          132.6981910970062,
          129.0302569977939,
          124.87670290283859,
          126.36769295204431
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:process-shared/sparse-skew/4",
      "language": "python",
      "backend": "process-shared",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 249.90877893287688,
        "q1": 218.40639697620645,
        "q3": 258.2219254691154,
        "min": 153.81875808816403,
        "max": 261.30698795896024,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 317.10223597474396,
        "q1": 290.12860101647675,
        "q3": 337.266226997599,
        "min": 263.15496605820954,
        "max": 357.43021802045405,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 793.9256140962243,
        "q1": 789.1623135074042,
        "q3": 819.370280019939,
        "min": 784.399012918584,
        "max": 844.8149459436536,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 2485.7867549872026,
        "q1": 2440.443523984868,
        "q3": 2551.4893514919095,
        "min": 2395.1002929825336,
        "max": 2617.1919479966164,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 82.2080078125,
      "max_processes": 6,
      "throughput_calls_per_second": {
        "median": 4.073868239598178,
        "q1": 3.877879666052284,
        "q3": 4.134534880792677,
        "min": 3.6818910925063904,
        "max": 4.195201521987177,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "process-shared",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 784.399012918584,
        "first_ms": 263.15496605820954,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 1227.3347359150648,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          216.1081830272451,
          194.42607800010592,
          206.03044400922954,
          258.1402739742771,
          258.3035769639537
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {},
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "CPython scalar loops",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "format": "d",
            "bytes": 16777216
          },
          "x": {
            "format": "d",
            "bytes": 262144
          },
          "output": {
            "format": "d",
            "bytes": 262144
          },
          "indptr": {
            "format": "Q",
            "bytes": 262152
          },
          "indices": {
            "format": "I",
            "bytes": 8388608
          }
        },
        "start_method": "spawn"
      }
    },
    {
      "id": "python:numpy-serial/dense/1",
      "language": "python",
      "backend": "numpy-serial",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.0523270359262824,
        "q1": 1.979848020710051,
        "q3": 2.550172503106296,
        "min": 1.6987570561468601,
        "max": 3.6569139920175076,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.0232169665396214,
        "q1": 2.683652506675571,
        "q3": 3.6661599879153073,
        "min": 2.344088046811521,
        "max": 4.309103009290993,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 158.60175096895546,
        "q1": 145.43248497648165,
        "q3": 162.8472349839285,
        "min": 132.26321898400784,
        "max": 167.09271899890155,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 237.57737700361758,
        "q1": 220.83450498757884,
        "q3": 239.28739147959277,
        "min": 204.0916329715401,
        "max": 240.99740595556796,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 56.0283203125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 309.7847365568377,
        "q1": 308.74206246624124,
        "q3": 329.5145172634954,
        "min": 307.69938837564473,
        "max": 349.2442979701531,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 132.26321898400784,
        "first_ms": 4.309103009290993,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 16.249626060016453,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          1.9885170040652156,
          1.9016950391232967,
          3.6569139920175076,
          1.9711790373548865,
          1.9938969053328037
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-serial/sparse-uniform/1",
      "language": "python",
      "backend": "numpy-serial",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.879404928535223,
        "q1": 2.299466053955257,
        "q3": 3.6551299854181707,
        "min": 2.0011940505355597,
        "max": 7.170509081333876,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.7411580085754395,
        "q1": 2.4604484788142145,
        "q3": 3.123142058029771,
        "min": 2.1797389490529895,
        "max": 3.5051261074841022,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 637.643311987631,
        "q1": 594.3442930001765,
        "q3": 679.7476254869252,
        "min": 551.0452740127221,
        "max": 721.8519389862195,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 836.95167908445,
        "q1": 785.1295510190539,
        "q3": 915.0281645124778,
        "min": 733.3074229536578,
        "max": 993.1046499405056,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 76.1591796875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 53.18139338537967,
        "q1": 50.7500198751279,
        "q3": 54.223298312806435,
        "min": 48.31864636487612,
        "max": 55.26520324023319,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 721.8519389862195,
        "first_ms": 3.5051261074841022,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 103.47972007002681,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          7.170509081333876,
          2.879404928535223,
          3.794496995396912,
          3.181873937137425,
          4.945135908201337
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-serial/sparse-skew/1",
      "language": "python",
      "backend": "numpy-serial",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.3122639395296574,
        "q1": 2.238933462649584,
        "q3": 2.9606865136884153,
        "min": 2.154362970031798,
        "max": 4.051134921610355,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 2.3798459442332387,
        "q1": 2.253359474707395,
        "q3": 2.832349971868098,
        "min": 2.126873005181551,
        "max": 3.284853999502957,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 571.8005510279909,
        "q1": 565.8150745439343,
        "q3": 692.2395284636877,
        "min": 559.8295980598778,
        "max": 812.6785058993846,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 761.7268890608102,
        "q1": 754.7973074833862,
        "q3": 897.1508190152235,
        "min": 747.8677259059623,
        "max": 1032.5747489696369,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 76.5263671875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 53.00093689710628,
        "q1": 52.915980654856355,
        "q3": 54.07720112134368,
        "min": 52.831024412606425,
        "max": 55.15346534558108,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 571.8005510279909,
        "first_ms": 3.284853999502957,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 94.64135998860002,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.002550966106355,
          2.154362970031798,
          2.653567004017532,
          2.6885049883276224,
          2.9729759553447366
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/dense/1",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.089384059421718,
        "q1": 2.13279394665733,
        "q3": 4.481845477130264,
        "min": 1.910642022266984,
        "max": 5.699811968952417,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.201080020517111,
        "q1": 2.8423755429685116,
        "q3": 3.9427754818461835,
        "min": 2.4836710654199123,
        "max": 4.684470943175256,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 132.09274294786155,
        "q1": 127.8269404429011,
        "q3": 167.79070999473333,
        "min": 123.56113793794066,
        "max": 203.48867704160511,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 204.6296939952299,
        "q1": 197.56235653767362,
        "q3": 261.6599124739878,
        "min": 190.49501908011734,
        "max": 318.6901309527457,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 56.0634765625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 270.73539378047417,
        "q1": 221.76546983439647,
        "q3": 276.8158225407593,
        "min": 172.7955458883188,
        "max": 282.8962513010445,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 132.09274294786155,
        "first_ms": 2.4836710654199123,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 18.468216992914677,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.219522932544351,
          2.2384029580280185,
          1.9990981090813875,
          2.9774309368804097,
          2.027184935286641
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-uniform/1",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.2438809974119067,
        "q1": 2.133273461367935,
        "q3": 2.803097479045391,
        "min": 1.844611018896103,
        "max": 3.869693959131837,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.6685270024463534,
        "q1": 3.6189969978295267,
        "q3": 4.004966001957655,
        "min": 3.5694669932127,
        "max": 4.3414050014689565,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 534.7539769718423,
        "q1": 482.6700474950485,
        "q3": 541.8228904600255,
        "min": 430.58611801825464,
        "max": 548.8918039482087,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 711.1910019302741,
        "q1": 649.1355469916016,
        "q3": 723.0831685010344,
        "min": 587.0800920529291,
        "max": 734.9753350717947,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 81.953125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 57.08686376350422,
        "q1": 56.49298387728202,
        "q3": 70.55508347312137,
        "min": 55.899103991059825,
        "max": 84.02330318273853,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 534.7539769718423,
        "first_ms": 3.5694669932127,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 87.58582395967096,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.7007380267605186,
          2.175802015699446,
          2.2438809974119067,
          2.1424139849841595,
          2.751133986748755
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-skew/1",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.962085069157183,
        "q1": 2.6714050327427685,
        "q3": 5.5261089582927525,
        "min": 2.1628899266943336,
        "max": 10.46097301878035,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.919568029232323,
        "q1": 3.1971500138752162,
        "q3": 8.587909513153136,
        "min": 2.4747319985181093,
        "max": 13.256250997073948,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 638.9409599360079,
        "q1": 580.7885169633664,
        "q3": 782.0491254678927,
        "min": 522.636073990725,
        "max": 925.1572909997776,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 841.1678980337456,
        "q1": 769.021061481908,
        "q3": 1053.8264365168288,
        "min": 696.8742249300703,
        "max": 1266.484974999912,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 76.8916015625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 51.92429783038309,
        "q1": 44.915812673407615,
        "q3": 54.98442108322119,
        "min": 37.90732751643214,
        "max": 58.044544336059296,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 925.1572909997776,
        "first_ms": 13.256250997073948,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 131.90061994828284,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.676489995792508,
          6.362528074532747,
          5.6359609588980675,
          5.5318999802693725,
          10.46097301878035
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131076
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/dense/2",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.139890031889081,
        "q1": 2.254478051327169,
        "q3": 3.8326375070028007,
        "min": 2.0655780099332333,
        "max": 4.914568038657308,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.964747000485659,
        "q1": 4.36541298404336,
        "q3": 5.883297533728182,
        "min": 3.766078967601061,
        "max": 6.801848066970706,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 154.99433397781104,
        "q1": 146.22009103186429,
        "q3": 162.60158852674067,
        "min": 137.44584808591753,
        "max": 170.2088430756703,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 236.31220299284905,
        "q1": 222.66061650589108,
        "q3": 251.92637200234458,
        "min": 209.00903001893312,
        "max": 267.5405410118401,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 56.0654296875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 278.0464783442774,
        "q1": 232.9173008895711,
        "q3": 278.1916169208282,
        "min": 187.7881234348648,
        "max": 278.3367554973791,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 154.99433397781104,
        "first_ms": 4.964747000485659,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 17.982605029828846,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.2307700710371137,
          2.278186031617224,
          3.358340007252991,
          2.9754230054095387,
          2.2811529925093055
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-uniform/2",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.8525729905813932,
        "q1": 3.192955511622131,
        "q3": 5.622247990686446,
        "min": 2.0761480554938316,
        "max": 9.302653023041785,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.3180210050195456,
        "q1": 3.4457980073057115,
        "q3": 5.929200502578169,
        "min": 2.5735750095918775,
        "max": 7.540380000136793,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 653.2785840099677,
        "q1": 602.0511200185865,
        "q3": 732.1356519823894,
        "min": 550.8236560272053,
        "max": 810.992719954811,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 951.92365499679,
        "q1": 843.7247160472907,
        "q3": 1034.1186100267805,
        "min": 735.5257770977914,
        "max": 1116.313565056771,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 81.9599609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 43.45508268762872,
        "q1": 41.79633458827536,
        "q3": 49.90205605496355,
        "min": 40.137586488922,
        "max": 56.349029422298386,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 550.8236560272053,
        "first_ms": 2.5735750095918775,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 88.73267297167331,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.2496269084513187,
          3.6985200131312013,
          3.8525729905813932,
          2.0761480554938316,
          2.3625590838491917
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-skew/2",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.4316110648214817,
        "q1": 2.299378509633243,
        "q3": 3.1053630518727005,
        "min": 2.085738000459969,
        "max": 3.8540909299626946,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.362608957104385,
        "q1": 3.0678485054522753,
        "q3": 3.919876995496452,
        "min": 2.7730880538001657,
        "max": 4.477145033888519,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 557.6248299330473,
        "q1": 539.3995049525984,
        "q3": 578.0275374418125,
        "min": 521.1741799721494,
        "max": 598.4302449505776,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 743.0426859064028,
        "q1": 722.9560819105245,
        "q3": 765.8896244829521,
        "min": 702.8694779146463,
        "max": 788.7365630595013,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 79.73046875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 57.00023733380816,
        "q1": 55.66432258405764,
        "q3": 57.39832519798536,
        "min": 54.328407834307136,
        "max": 57.79641306216256,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 598.4302449505776,
        "first_ms": 2.7730880538001657,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 86.51055896189064,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.085738000459969,
          3.0943240271881223,
          2.2565360413864255,
          2.3360539926216006,
          3.1164020765572786
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131080
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/dense/4",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.3779629739001393,
        "q1": 2.5871825055219233,
        "q3": 3.6031469935551286,
        "min": 1.9063029903918505,
        "max": 4.196516005322337,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.639616003260016,
        "q1": 4.253590013831854,
        "q3": 5.069237493444234,
        "min": 3.8675640244036913,
        "max": 5.498858983628452,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 156.23073489405215,
        "q1": 141.0113544552587,
        "q3": 159.32976093608886,
        "min": 125.79197401646525,
        "max": 162.42878697812557,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 237.61772899888456,
        "q1": 216.11668047262356,
        "q3": 246.09371152473614,
        "min": 194.61563194636256,
        "max": 254.5696940505877,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 56.103515625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 258.24620199469445,
        "q1": 238.81291489381675,
        "q3": 259.6197504771958,
        "min": 219.3796277929391,
        "max": 260.9932989596971,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 125.79197401646525,
        "first_ms": 3.8675640244036913,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 19.157579983584583,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.3779629739001393,
          2.769788959994912,
          3.553098998963833,
          2.8984000673517585,
          1.9063029903918505
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "NumPy dot; BLAS threads=1",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-uniform/4",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.1160860089585185,
        "q1": 2.4949859944172204,
        "q3": 3.321158525068313,
        "min": 2.156652044504881,
        "max": 3.981032990850508,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 3.6290899151936173,
        "q1": 3.544471983332187,
        "q3": 3.7221434176899493,
        "min": 3.4598540514707565,
        "max": 3.815196920186281,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 552.9181539313868,
        "q1": 542.8272574790753,
        "q3": 557.3108739918098,
        "min": 532.7363610267639,
        "max": 561.7035940522328,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 742.8843819070607,
        "q1": 734.3576929415576,
        "q3": 746.1895274464041,
        "min": 725.8310039760545,
        "max": 749.4946729857475,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 100.599609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 53.5731720016593,
        "q1": 53.44513397402912,
        "q3": 54.36446767784602,
        "min": 53.317095946398936,
        "max": 55.15576335403274,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 561.7035940522328,
        "first_ms": 3.815196920186281,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.33029598928988,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.981032990850508,
          3.1205270206555724,
          3.3087250776588917,
          2.1830840269103646,
          2.8252010233700275
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numpy-thread/sparse-skew/4",
      "language": "python",
      "backend": "numpy-thread",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.984542050398886,
        "q1": 2.5099809863604605,
        "q3": 3.3970099757425487,
        "min": 2.3493269691243768,
        "max": 3.8370470283553004,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 4.886450013145804,
        "q1": 4.259990993887186,
        "q3": 5.077533016446978,
        "min": 3.6335319746285677,
        "max": 5.268616019748151,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 552.385191898793,
        "q1": 547.5999754853547,
        "q3": 559.8941914504394,
        "min": 542.8147590719163,
        "max": 567.4031910020858,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 742.9061270086095,
        "q1": 735.1395349833183,
        "q3": 753.2269805087708,
        "min": 727.3729429580271,
        "max": 763.5478340089321,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 81.4599609375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 53.56540953829262,
        "q1": 53.14736265735327,
        "q3": 54.85321713121502,
        "min": 52.72931577641393,
        "max": 56.14102472413741,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numpy-thread",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 552.385191898793,
        "first_ms": 5.268616019748151,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.34382100496441,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.6813920121639967,
          2.984542050398886,
          2.995487069711089,
          3.7618849892169237,
          2.5313899386674166
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "scipy": "1.15.3"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "SciPy CSR matvec; contiguous row slabs",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "int32",
            "bytes": 131076
          },
          "indices": {
            "dtype": "int32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 131088
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numba-serial/dense/1",
      "language": "python",
      "backend": "numba-serial",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.37041197437793,
        "q1": 3.966497490182519,
        "q3": 5.711562524084002,
        "min": 2.8674990171566606,
        "max": 11.198604013770819,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 337.81525294762105,
        "q1": 297.2508204402402,
        "q3": 342.659312475007,
        "min": 256.68638793285936,
        "max": 347.503372002393,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 288.7423620559275,
        "q1": 287.5452675507404,
        "q3": 289.0879045589827,
        "min": 286.34817304555327,
        "max": 289.4334470620379,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 884.2414440587163,
        "q1": 865.1399959926493,
        "q3": 941.47579401033,
        "min": 846.0385479265824,
        "max": 998.7101439619437,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 171.81640625,
      "max_processes": 2,
      "throughput_calls_per_second": {
        "median": 165.70316375890894,
        "q1": 165.64970010429263,
        "q3": 169.688721669786,
        "min": 165.59623644967633,
        "max": 173.67427958066307,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 286.34817304555327,
        "first_ms": 337.81525294762105,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 30.17443895805627,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.1274619763717055,
          3.7540659541264176,
          6.925056921318173,
          5.083882017061114,
          4.217181005515158
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numba-serial/sparse-uniform/1",
      "language": "python",
      "backend": "numba-serial",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.707791980355978,
        "q1": 1.9892179989255965,
        "q3": 3.5134819918312132,
        "min": 1.880978001281619,
        "max": 4.086204920895398,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 336.6569150239229,
        "q1": 326.8908294849098,
        "q3": 436.5480955457315,
        "min": 317.1247439458966,
        "max": 536.43927606754,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 608.1609459361061,
        "q1": 587.9462399752811,
        "q3": 720.0264589628205,
        "min": 567.731534014456,
        "max": 831.8919719895348,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1242.694205022417,
        "q1": 1212.1269369963557,
        "q3": 1496.6228344710544,
        "min": 1181.5596689702943,
        "max": 1750.5514639196917,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 162.8955078125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 55.325718367993936,
        "q1": 53.58462554105604,
        "q3": 56.70945553952909,
        "min": 51.843532714118155,
        "max": 58.09319271106424,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 608.1609459361061,
        "first_ms": 336.6569150239229,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 90.37388302385807,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.001799992285669,
          3.3614039421081543,
          2.4614569265395403,
          1.956893946044147,
          1.976636005565524
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numba-serial/sparse-skew/1",
      "language": "python",
      "backend": "numba-serial",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.833493985235691,
        "q1": 2.403659513220191,
        "q3": 5.971928010694683,
        "min": 2.0669150399044156,
        "max": 11.277138954028487,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 356.123381992802,
        "q1": 336.31981699727476,
        "q3": 488.4690579492599,
        "min": 316.5162520017475,
        "max": 620.8147339057177,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 592.4599739955738,
        "q1": 568.6703730025329,
        "q3": 781.2071484513581,
        "min": 544.8807720094919,
        "max": 969.9543229071423,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1253.1604469986632,
        "q1": 1207.6220154995099,
        "q3": 1706.9883209769614,
        "min": 1162.0835840003565,
        "max": 2160.8161949552596,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 162.9560546875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 51.17133021078773,
        "q1": 44.588453356412856,
        "q3": 54.121154263889025,
        "min": 38.00557650203799,
        "max": 57.07097831699032,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-serial",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 969.9543229071423,
        "first_ms": 620.8147339057177,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 131.55964098405093,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          11.277138954028487,
          8.79319105297327,
          6.735525093972683,
          5.208330927416682,
          6.848598946817219
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/dense/1",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 1,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 5.167854949831963,
        "q1": 3.7556500174105167,
        "q3": 5.531804985366762,
        "min": 3.5980320535600185,
        "max": 7.86728598177433,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 603.4000850049779,
        "q1": 586.2868304830045,
        "q3": 656.5850609913468,
        "min": 569.173575961031,
        "max": 709.7700369777158,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 315.6536449678242,
        "q1": 306.88558745896444,
        "q3": 331.85661694733426,
        "min": 298.11752995010465,
        "max": 348.0595889268443,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1161.474140943028,
        "q1": 1124.1698989761062,
        "q3": 1249.8197484528646,
        "min": 1086.8656570091844,
        "max": 1338.1653559627011,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 172.501953125,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 188.27892872533124,
        "q1": 164.00310816536904,
        "q3": 190.6103508850676,
        "min": 139.72728760540681,
        "max": 192.94177304480394,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 348.0595889268443,
        "first_ms": 709.7700369777158,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 35.78399098478258,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.742968991398811,
          5.555578973144293,
          5.9742790181189775,
          7.86728598177433,
          5.5080309975892305
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-uniform/1",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 1,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.8810410294681787,
        "q1": 2.5900555192492902,
        "q3": 3.1680590473115444,
        "min": 2.142398036085069,
        "max": 5.216371035203338,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 654.0229379897937,
        "q1": 589.9074019980617,
        "q3": 841.4537155185826,
        "min": 525.7918660063297,
        "max": 1028.8844930473715,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 668.8254770124331,
        "q1": 620.7898674765602,
        "q3": 796.1140415281989,
        "min": 572.7542579406872,
        "max": 923.4026060439646,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1660.633870982565,
        "q1": 1535.6823269394226,
        "q3": 1986.802312952932,
        "min": 1410.7307828962803,
        "max": 2312.970754923299,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.0732421875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 55.080201198850034,
        "q1": 53.80679074349314,
        "q3": 56.26906438627905,
        "min": 52.53338028813625,
        "max": 57.45792757370806,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 572.7542579406872,
        "first_ms": 525.7918660063297,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 95.17757990397513,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.430350985378027,
          2.463288023136556,
          2.6100960094481707,
          2.57001502905041,
          3.1736460514366627
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-skew/1",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 1,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.9349590186029673,
        "q1": 2.8221054817549884,
        "q3": 3.351013467181474,
        "min": 2.113310038112104,
        "max": 4.0362890576943755,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 699.3943839333951,
        "q1": 628.0601234757341,
        "q3": 706.1062819557264,
        "min": 556.7258630180731,
        "max": 712.8181799780577,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 676.9512370228767,
        "q1": 637.3787054908462,
        "q3": 679.8508894862607,
        "min": 597.8061739588156,
        "max": 682.7505419496447,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1733.6387620307505,
        "q1": 1592.8379995166324,
        "q3": 1750.170434010215,
        "min": 1452.0372370025143,
        "max": 1766.7021059896797,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.0556640625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 51.63627185086633,
        "q1": 51.29918190410062,
        "q3": 52.9768359693083,
        "min": 50.96209195733492,
        "max": 54.31740008775027,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 1,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 676.9512370228767,
        "first_ms": 712.8181799780577,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 98.11214194633067,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.9349590186029673,
          3.594570909626782,
          4.0362890576943755,
          3.404259914532304,
          3.297767019830644
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 1,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 1,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/dense/2",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 2,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.1585080325603485,
        "q1": 3.603706485591829,
        "q3": 4.448423511348665,
        "min": 3.4836989361792803,
        "max": 5.269601009786129,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 537.6138689462095,
        "q1": 537.5231514917687,
        "q3": 546.8109709909186,
        "min": 537.4324340373278,
        "max": 556.0080730356276,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 284.4082290539518,
        "q1": 280.2397125051357,
        "q3": 285.51034053089097,
        "min": 276.07119595631957,
        "max": 286.61245200783014,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1051.2462359620258,
        "q1": 1036.0977469827048,
        "q3": 1068.5889724991284,
        "min": 1020.9492580033839,
        "max": 1085.931709036231,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 172.560546875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 202.31706458889337,
        "q1": 195.3234774724917,
        "q3": 202.7012259412779,
        "min": 188.32989035609003,
        "max": 203.08538729366242,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 276.07119595631957,
        "first_ms": 537.4324340373278,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 24.713683989830315,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.721665940247476,
          4.291919060051441,
          3.510736976750195,
          5.269601009786129,
          3.4836989361792803
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-uniform/2",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 2,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 3.1620230292901397,
        "q1": 2.7965829940512776,
        "q3": 7.216509024146944,
        "min": 2.114937989972532,
        "max": 11.09870697837323,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 700.1412380486727,
        "q1": 632.5181830325164,
        "q3": 842.9404320195317,
        "min": 564.8951280163601,
        "max": 985.7396259903908,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 720.6852249801159,
        "q1": 651.2732214760035,
        "q3": 789.5393239450641,
        "min": 581.8612179718912,
        "max": 858.3934229100123,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1897.0726950792596,
        "q1": 1677.6633340632543,
        "q3": 2162.938619032502,
        "min": 1458.253973047249,
        "max": 2428.804542985745,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.115234375,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 46.311191346019804,
        "q1": 41.551524683142105,
        "q3": 49.848177638741234,
        "min": 36.79185802026441,
        "max": 53.385163931462664,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 581.8612179718912,
        "first_ms": 564.8951280163601,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 93.65897998213768,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.4171960540115833,
          2.133582951501012,
          2.720469026826322,
          2.114937989972532,
          2.904004999436438
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-skew/2",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 2,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.823401940986514,
        "q1": 2.3822050425224006,
        "q3": 3.5875444882549345,
        "min": 2.061723032966256,
        "max": 5.423299968242645,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 543.7732299324125,
        "q1": 535.4212339734659,
        "q3": 551.7715289606713,
        "min": 527.0692380145192,
        "max": 559.76982798893,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 564.6512450184673,
        "q1": 554.6378585277125,
        "q3": 598.4486340312287,
        "min": 544.6244720369577,
        "max": 632.2460230439901,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1445.7830809988081,
        "q1": 1408.6681515327655,
        "q3": 1491.8547694687732,
        "min": 1371.5532220667228,
        "max": 1537.9264579387382,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.0419921875,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 55.551314573084305,
        "q1": 53.80880757914947,
        "q3": 56.153812800906536,
        "min": 52.06630058521464,
        "max": 56.756311028728774,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 2,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 632.2460230439901,
        "first_ms": 559.76982798893,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 96.03140503168106,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          5.423299968242645,
          3.8739809533581138,
          2.158384071663022,
          2.823401940986514,
          3.37814399972558
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 2,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 2,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/dense/4",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 4,
      "workload": "dense",
      "rows": 2048,
      "cols": 2048,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 4.799041897058487,
        "q1": 4.406159452628344,
        "q3": 5.55877952137962,
        "min": 2.8762080473825336,
        "max": 8.536998997442424,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 694.3662739358842,
        "q1": 645.3181389370002,
        "q3": 843.6145384330302,
        "min": 596.2700039381161,
        "max": 992.8628029301763,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 340.7541300402954,
        "q1": 337.7597145154141,
        "q3": 429.3584005208686,
        "min": 334.76529899053276,
        "max": 517.9626710014418,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1302.3972290102392,
        "q1": 1256.404690502677,
        "q3": 1551.5624194522388,
        "min": 1210.412151995115,
        "max": 1800.7276098942384,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 172.634765625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 171.0991209174941,
        "q1": 168.51874764703092,
        "q3": 173.82571391530024,
        "min": 165.93837437656777,
        "max": 176.5523069131064,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "dense",
        "rows": 2048,
        "cols": 2048,
        "nnz_per_row": 64,
        "nnz": 4194304,
        "repeats": 5,
        "setup_ms": 517.9626710014418,
        "first_ms": 992.8628029301763,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 28.320219018496573,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          4.279211978428066,
          8.536998997442424,
          2.921594073995948,
          6.816543987952173,
          2.8762080473825336
        ],
        "checksum": "41338",
        "output_hash": "0db0f71d8f68135f",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 33587200,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 33554432
          },
          "x": {
            "dtype": "float64",
            "bytes": 16384
          },
          "output": {
            "dtype": "float64",
            "bytes": 16384
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-uniform/4",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 4,
      "workload": "sparse-uniform",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.5311369681730866,
        "q1": 2.1168544772081077,
        "q3": 3.240937483496964,
        "min": 2.010046038776636,
        "max": 8.060817955993116,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 531.9862050237134,
        "q1": 524.0931825246662,
        "q3": 645.2126000076532,
        "min": 516.200160025619,
        "max": 758.4389949915931,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 559.5933699514717,
        "q1": 551.0775659931824,
        "q3": 592.7293485146947,
        "min": 542.561762034893,
        "max": 625.8653270779178,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1374.7666700510308,
        "q1": 1365.7687405357137,
        "q3": 1637.6490275142714,
        "min": 1356.7708110203966,
        "max": 1900.531384977512,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.275390625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 55.6048764193379,
        "q1": 48.00203790463998,
        "q3": 56.840207020578546,
        "min": 40.39919938994205,
        "max": 58.07553762181919,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-uniform",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 542.561762034893,
        "first_ms": 516.200160025619,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.92017107084394,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          2.073963056318462,
          2.733201952651143,
          2.0173259545117617,
          2.9357189778238535,
          2.5311369681730866
        ],
        "checksum": "7278",
        "output_hash": "18cc5ed244ef97d9",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    },
    {
      "id": "python:numba-parallel/sparse-skew/4",
      "language": "python",
      "backend": "numba-parallel",
      "workers": 4,
      "workload": "sparse-skew",
      "rows": 32768,
      "cols": 32768,
      "nnz_per_row": 64,
      "warm_ms": {
        "median": 2.8166320407763124,
        "q1": 2.666652959305793,
        "q3": 2.985519473440945,
        "min": 2.144285012036562,
        "max": 3.2981010153889656,
        "p95": null,
        "p99": null,
        "count": 15
      },
      "first_ms": {
        "median": 675.0377389835194,
        "q1": 602.3000379791483,
        "q3": 684.2424689675681,
        "min": 529.5623369747773,
        "max": 693.4471989516169,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "setup_ms": {
        "median": 712.462816038169,
        "q1": 640.5671590473503,
        "q3": 715.3767730342224,
        "min": 568.6715020565316,
        "max": 718.2907300302759,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "process_wall_ms": {
        "median": 1720.9215310867876,
        "q1": 1554.2271530721337,
        "q3": 1731.7693200311624,
        "min": 1387.5327750574797,
        "max": 1742.6171089755371,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "peak_tree_pss_mib": 165.197265625,
      "max_processes": 1,
      "throughput_calls_per_second": {
        "median": 54.80392462377968,
        "q1": 54.16494661217001,
        "q3": 55.23890285257855,
        "min": 53.52596860056035,
        "max": 55.67388108137741,
        "p95": null,
        "p99": null,
        "count": 3
      },
      "measurement_details": {
        "backend": "numba-parallel",
        "language": "python",
        "workers": 4,
        "workload": "sparse-skew",
        "rows": 32768,
        "cols": 32768,
        "nnz_per_row": 64,
        "nnz": 2097152,
        "repeats": 5,
        "setup_ms": 568.6715020565316,
        "first_ms": 529.5623369747773,
        "stress_seconds_requested": 0.0,
        "elapsed_warm_ms": 89.80872004758567,
        "iterations": 5,
        "validation_scope": "all",
        "samples_ms": [
          3.1526120146736503,
          2.629810944199562,
          2.302483073435724,
          2.163612050935626,
          2.144285012036562
        ],
        "checksum": "538",
        "output_hash": "42c390bd74645f49",
        "validated": true,
        "python": "3.12.3",
        "versions": {
          "numpy": "2.2.6",
          "numba": "0.61.2"
        },
        "python_implementation": "CPython",
        "python_compiler": "GCC 13.3.0",
        "python_full": "3.12.3 (main, Aug 31 2026, 10:18:26) [GCC 13.3.0]",
        "implementation": "Numba fused row loops; fastmath=False",
        "partition": "equal contiguous rows",
        "logical_data_bytes": 25952264,
        "storage": {
          "values": {
            "dtype": "float64",
            "bytes": 16777216
          },
          "x": {
            "dtype": "float64",
            "bytes": 262144
          },
          "output": {
            "dtype": "float64",
            "bytes": 262144
          },
          "indptr": {
            "dtype": "uint64",
            "bytes": 262152
          },
          "indices": {
            "dtype": "uint32",
            "bytes": 8388608
          },
          "shard_row_pointer_bytes": 0
        },
        "threading_layer": "omp",
        "numba_threads": 4,
        "threadpool_runtimes": [
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.29",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "blas",
            "internal_api": "openblas",
            "num_threads": 1,
            "prefix": "libscipy_openblas",
            "version": "0.3.28",
            "threading_layer": "pthreads",
            "architecture": "Haswell"
          },
          {
            "user_api": "openmp",
            "internal_api": "openmp",
            "num_threads": 4,
            "prefix": "libgomp",
            "version": null
          }
        ]
      }
    }
  ]
}
