{"id":20433766,"url":"https://github.com/intelpython/composability_bench","last_synced_at":"2025-05-08T17:33:27.458Z","repository":{"id":66109070,"uuid":"82825260","full_name":"IntelPython/composability_bench","owner":"IntelPython","description":"Show effects of over-subscription and ways to fix that","archived":false,"fork":false,"pushed_at":"2024-08-15T17:56:00.000Z","size":64,"stargazers_count":15,"open_issues_count":3,"forks_count":5,"subscribers_count":5,"default_branch":"master","last_synced_at":"2024-08-16T16:05:01.334Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","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/IntelPython.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":"2017-02-22T16:14:05.000Z","updated_at":"2024-08-15T17:56:02.000Z","dependencies_parsed_at":"2024-08-15T15:59:12.277Z","dependency_job_id":null,"html_url":"https://github.com/IntelPython/composability_bench","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IntelPython%2Fcomposability_bench","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IntelPython%2Fcomposability_bench/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IntelPython%2Fcomposability_bench/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IntelPython%2Fcomposability_bench/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/IntelPython","download_url":"https://codeload.github.com/IntelPython/composability_bench/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224749751,"owners_count":17363554,"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":[],"created_at":"2024-11-15T08:20:59.303Z","updated_at":"2024-11-15T08:20:59.777Z","avatar_url":"https://github.com/IntelPython.png","language":"Python","readme":"# Composability Benchmarks\nShow effects of oversubscription and ways to fix that.\n\n## Installation\nInstall TBB and SMP modules for Python in order to evaluate composable multithreading:\n```\nconda install -c intel tbb4py smp\n```\nIf not sure how to do this, just run `set_python_envs.sh` to set up conda environment with all the necessary components of Intel Distribution for Python and follow instructions for environment activation, e.g. `conda activate intel3`.\n\n## Running\nEffects are visible on big enough machine with 32 and more cores.\nRun following modes for example:\n\n```\npython collab_filt.py\npython dask_bench2.py\npython -m tbb collab_filt.py\npython -m tbb dask_bench2.py\npython -m smp collab_filt.py\npython -m smp dask_bench2.py\n```\n\n## Composability Modes\nThere are the folloing composability modes for testing:\n\n### `-m tbb`\nEnables TBB threading for MKL, Numpy, Dask, Python's multiprocessing.ThreadPool\n\n### `-m tbb --ipc`\nSame as `-m tbb` but also enables interprocess coordination for multiprocessing applications.\n\n### `-m smp`\nStatically allocates CPU resources between the nested parallel regions using affinity masks and OpenMP API. Supports both multithreading and multiprocessing parallelism.\n\n### `-m smp -o`\nEnables `KMP_COMPOSABILITY=mode=counting` for Intel OpenMP runtime when parallel regions are ordered using a semaphore. Supports both multithreading and multiprocessing parallelism.\n\n## See also\n\nPaper: \"[Composable Multi-Threading and Multi-Processing for Numeric Libraries](http://conference.scipy.org/proceedings/scipy2018/pdfs/anton_malakhov.pdf)\" by Anton Malakhov, David Liu, Anton Gorshkov, Terry Wilmarth. Proceedings of the 17th Python in Science Conference (SciPy 2018), Austin, Texas (July 9 - 15, 2018). DOI 10.25080/Majora-4af1f417-003\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fintelpython%2Fcomposability_bench","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fintelpython%2Fcomposability_bench","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fintelpython%2Fcomposability_bench/lists"}