{"id":17360724,"url":"https://github.com/certik/matmul","last_synced_at":"2026-01-06T22:47:49.583Z","repository":{"id":167367933,"uuid":"642644109","full_name":"certik/matmul","owner":"certik","description":"Matrix multiplication benchmark","archived":false,"fork":false,"pushed_at":"2023-11-29T16:56:24.000Z","size":21,"stargazers_count":5,"open_issues_count":6,"forks_count":2,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-02-01T20:13:32.546Z","etag":null,"topics":["benchmark","fortran","matrix-multiplication"],"latest_commit_sha":null,"homepage":"","language":"Fortran","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/certik.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":"2023-05-19T03:09:18.000Z","updated_at":"2023-09-10T18:15:22.000Z","dependencies_parsed_at":null,"dependency_job_id":"542684ed-8eed-4254-bc84-86e265973429","html_url":"https://github.com/certik/matmul","commit_stats":null,"previous_names":["certik/matmul"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/certik%2Fmatmul","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/certik%2Fmatmul/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/certik%2Fmatmul/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/certik%2Fmatmul/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/certik","download_url":"https://codeload.github.com/certik/matmul/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245881627,"owners_count":20687754,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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","fortran","matrix-multiplication"],"created_at":"2024-10-15T19:27:44.599Z","updated_at":"2026-01-06T22:47:49.544Z","avatar_url":"https://github.com/certik.png","language":"Fortran","readme":"# MatMul Benchmark\n\nRun:\n```\nFC=gfortran cmake -DMATMUL_BLAS=OpenBLAS .\nmake\nOMP_NUM_THREADS=1 ./matmul\n```\n\n# Benchmark results on Apple M1\n\nThe theoretical performance peak for matmul is just the cost of `fma`, which is\n0.125 clock cycles per double precision matrix element (`fmla.2d v0, v0, v0`\ntakes 0.25 cycles), and 0.0625 per single precision element.\n\nSingle precison (f32) matmul\n\npeak = 0.0625 clock cycles\n\n    n    OpenBlas\n    512  0.0768\n    1024 0.0672\n    2048 0.0640\n    4096 0.0632\n    8192 0.0631\n\nTo convert these clock cycles to seconds, multiply by n^3 and divide by 3.2GHz.\nFor example the n=8192 case gives 10.84s:\n\n    \u003e\u003e\u003e n = 8192; 0.0631*n**3 / 3.2e9\n    10.840497455104\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcertik%2Fmatmul","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcertik%2Fmatmul","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcertik%2Fmatmul/lists"}