{"id":18698125,"url":"https://github.com/nickolasrm/gpuvscpumatrixmultiplication","last_synced_at":"2025-09-06T04:44:45.652Z","repository":{"id":103753556,"uuid":"192747531","full_name":"nickolasrm/GPUvsCPUMatrixMultiplication","owner":"nickolasrm","description":"CPU and GPU optimized matrix multiplication (AVX, transposition, CUDA and other)","archived":false,"fork":false,"pushed_at":"2020-09-09T04:45:42.000Z","size":400,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-05-19T01:37:48.453Z","etag":null,"topics":["avx","comparison","cuda","hpc","matrix","multiplication"],"latest_commit_sha":null,"homepage":"","language":"Cuda","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/nickolasrm.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":"2019-06-19T14:23:07.000Z","updated_at":"2023-08-21T08:38:12.000Z","dependencies_parsed_at":"2023-03-18T22:49:15.959Z","dependency_job_id":null,"html_url":"https://github.com/nickolasrm/GPUvsCPUMatrixMultiplication","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/nickolasrm/GPUvsCPUMatrixMultiplication","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickolasrm%2FGPUvsCPUMatrixMultiplication","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickolasrm%2FGPUvsCPUMatrixMultiplication/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickolasrm%2FGPUvsCPUMatrixMultiplication/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickolasrm%2FGPUvsCPUMatrixMultiplication/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nickolasrm","download_url":"https://codeload.github.com/nickolasrm/GPUvsCPUMatrixMultiplication/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nickolasrm%2FGPUvsCPUMatrixMultiplication/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273858864,"owners_count":25180771,"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","status":"online","status_checked_at":"2025-09-06T02:00:13.247Z","response_time":2576,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["avx","comparison","cuda","hpc","matrix","multiplication"],"created_at":"2024-11-07T11:27:06.532Z","updated_at":"2025-09-06T04:44:45.621Z","avatar_url":"https://github.com/nickolasrm.png","language":"Cuda","funding_links":[],"categories":[],"sub_categories":[],"readme":"# GPU vs CPU analysis\nWhat has been made:\n- Transposition\n- Optimization of recurrent operations in loops\n- AVX\n- OpenMP\n- CUDA\n- Tile based\n\n**Usage**\n- Compile it by using \"make\"\n- Generate a matrix with \"./BINARY g L1 C1 L2 C2 s\"\n  - \"BINARY\" is the binary to be used (./cuda or ./norm)\n  - \"g\" tells the software to generate a matrix\n  - \"L1 C1\" are respectively matrix A lines and columns\n  - \"L2 C2\" are respectively matrix B lines and columns\n  - \"s\", (optional) tells the software to save matrix A and B\n- Reading a matrix from file \"./BINARY f L1 C1 L2 C2 FILE1 FILE2\"\n  - \"BINARY\" is the binary to be used (./cuda or ./norm)\n  - \"f\" tells the software to read a matrix from a file\n  - \"L1 C1\" are respectively matrix A lines and columns\n  - \"L2 C2\" are respectively matrix B lines and columns\n  - \"FILE1\" is the name of matrix A file\n  - \"FILE2\" is the name of matrix B file\n- Compile matrix comparator with \"make comparator\"\n- Compare matrix multiplication results with  \"./comp C1.txt C2.txt\"\n- Rename matrices from 0....txt, 1....txt e  2....txt to a.txt, b.txt and c.txt with \"make rename\"\n- Clean txts with \"make clean\"\n\n**NOTE**\n### Source files\n- matrixCuda.cu: Optimized matrix multiplication (CUDA, AVX, Transposition...)\n- matrix.cu: Not optmized matrix multiplication\n- comparator.c: Matrix equality file comparator\n\n### Generated binaries\n- cuda: Optimized matrix multiplication (CUDA, AVX, Transposition...)\n- norm: Not optimized matrix multiplication\n- comp: Matrix equality file comparator\n\n### Output files saved with 's' flag:\n- Matrix A: 0-rowsxcols.txt\n- Matrix B: 1-rowsxcols.txt\n- Matrix C: 2-rowsxcols.txt\n\nWhether executing the software with the same matrix dimensions, the output files will be overwritten.\nThus, is better to use \"make rename\" to avoid it.\n\n### Usage example:\n```\n./cuda g 1000 1000 1000 1000 s\nmake rename\n./norm f 1000 1000 1000 1000 a.txt b.txt\n./comp 0-1000x1000.txt c.txt\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnickolasrm%2Fgpuvscpumatrixmultiplication","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnickolasrm%2Fgpuvscpumatrixmultiplication","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnickolasrm%2Fgpuvscpumatrixmultiplication/lists"}