{"id":23555559,"url":"https://github.com/kostrykin/blas-benchmark","last_synced_at":"2026-05-04T20:33:04.252Z","repository":{"id":224086238,"uuid":"762350002","full_name":"kostrykin/blas-benchmark","owner":"kostrykin","description":"A benchmark comparison of different BLAS backends for NumPy.","archived":false,"fork":false,"pushed_at":"2024-02-26T19:53:36.000Z","size":136,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-07-23T18:53:16.780Z","etag":null,"topics":["benchmark","blas","cvxpy","mkl","numpy","openblas","python"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/kostrykin.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-02-23T15:42:25.000Z","updated_at":"2024-02-24T14:21:24.000Z","dependencies_parsed_at":"2024-12-26T13:19:12.484Z","dependency_job_id":"b5f5cee4-820a-4ea3-9e06-f1ba3a5fb2a2","html_url":"https://github.com/kostrykin/blas-benchmark","commit_stats":null,"previous_names":["kostrykin/blas-benchmark"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/kostrykin/blas-benchmark","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kostrykin%2Fblas-benchmark","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kostrykin%2Fblas-benchmark/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kostrykin%2Fblas-benchmark/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kostrykin%2Fblas-benchmark/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/kostrykin","download_url":"https://codeload.github.com/kostrykin/blas-benchmark/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kostrykin%2Fblas-benchmark/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32624143,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-04T10:08:07.713Z","status":"ssl_error","status_checked_at":"2026-05-04T10:08:02.005Z","response_time":58,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["benchmark","blas","cvxpy","mkl","numpy","openblas","python"],"created_at":"2024-12-26T13:18:46.020Z","updated_at":"2026-05-04T20:33:04.216Z","avatar_url":"https://github.com/kostrykin.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# blas-benchmark\n\nTasks are defined in `tasks/*.py`. The Conda environments, which specify the different BLAS configurations, are defined in `results/*/environment.yml`. Python versions and numbers of threads are defined in `profiles.yml`.\n\nTo determine the runtime of a task, each task is repeated for at least 10 seconds, and the average is determined. The repetition and averaging procedure is repeated 3 times, and the best result is used.\n\n## Main results:\n\nThe configurations \u003ccode\u003emkl2020.0_debug\u003c/code\u003e and \u003ccode\u003emkl2020.1_fakeintel\u003c/code\u003e perform overall best:\n\n\u003ctable\u003e\n  \u003ctr\u003e\n    \u003cth rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\n    \u003cth colspan=\"2\"\u003eAMD Ryzen Threadripper 3970X\u003c/th\u003e\n    \u003cth\u003eAMD EPYC 7763\u003c/th\u003e\n  \u003ctr\u003e\n    \u003ctd\u003e\u003ca href=\"https://github.com/kostrykin/blas-benchmark/blob/master/reports/py38_2threads/AMD%20Ryzen%20Threadripper%203970X%2032-Core%20Processor.ipynb\"\u003e2 threads\u003c/a\u003e\u003c/td\u003e\n    \u003ctd\u003e\u003ca href=\"https://github.com/kostrykin/blas-benchmark/blob/master/reports/py38_16threads/AMD%20Ryzen%20Threadripper%203970X%2032-Core%20Processor.ipynb\"\u003e16 threads\u003c/a\u003e\u003c/td\u003e\n    \u003ctd\u003e\u003ca href=\"https://github.com/kostrykin/blas-benchmark/blob/master/reports/py38_2threads/AMD%20EPYC%207763%2064-Core%20Processor.ipynb\"\u003e2 threads\u003c/a\u003e\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n    \u003ctd\u003e\u003ccode\u003eopenblas\u003c/code\u003e\u003c/td\u003e\n    \u003ctd\u003e1.003829\u003c/td\u003e\n    \u003ctd\u003e1.019895\u003c/td\u003e\n    \u003ctd\u003e1.016262\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n    \u003ctd\u003e\u003ccode\u003emkl2024.0\u003c/code\u003e\u003c/td\u003e\n    \u003ctd\u003e1.128423\u003c/td\u003e\n    \u003ctd\u003e1.213864\u003c/td\u003e\n    \u003ctd\u003e1.065984\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n    \u003ctd\u003e\u003ccode\u003emkl2020.0_debug\u003c/code\u003e\u003c/td\u003e\n    \u003ctd\u003e\u003cb\u003e1.156737\u003c/b\u003e\u003c/td\u003e\n    \u003ctd\u003e1.261223\u003c/td\u003e\n    \u003ctd\u003e1.162273\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n    \u003ctd\u003e\u003ccode\u003emkl2020.1_fakeintel\u003c/code\u003e\u003c/td\u003e\n    \u003ctd\u003e1.144065\u003c/td\u003e\n    \u003ctd\u003e\u003cb\u003e1.281782\u003c/b\u003e\u003c/td\u003e\n    \u003ctd\u003e\u003cb\u003e1.164156\u003c/b\u003e\u003c/td\u003e\n  \u003c/tr\u003e\n\u003c/table\u003e\n\nThe *score* of a configuration is the *geometric mean* of the best possible speed-up in comparison to the other configurations. See `reports/*.ipynb` for details.\n\n## Benchmark CLI:\n\nRun the benchmark on your CPU:\n```\npython -m benchmark.cli --profiles py38_2threads py38_16threads --run\n```\n\nOr only update the reports:\n```\npython -m benchmark.cli\n```\n\n## Acknowledgements:\n- \u003chttps://gist.github.com/cty-yyds/41bcfa6a71670527c93049aa9a5d249f\u003e\n- \u003chttps://gist.github.com/bebosudo/6f43dc6b4329c197f258f25cc69f0ec0\u003e\n- \u003chttps://danieldk.eu/Posts/2020-08-31-MKL-Zen.html\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkostrykin%2Fblas-benchmark","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkostrykin%2Fblas-benchmark","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkostrykin%2Fblas-benchmark/lists"}