.. image:: https://github.com/RRZE-HPC/kerncraft/blob/master/doc/logo/logo-lightbg.svg
kerncraft
=========
Loop Kernel Analysis and Performance Modeling Toolkit
This tool allows automatic analysis of loop kernels using the `Execution Cache Memory (ECM) model <https://hpc.fau.de/research/ecm/>`_,
the Roofline model and actual benchmarks. kerncraft provides a framework to investigate the
data reuse and cache requirements by static code analysis. In combination with the Intel IACA tool
kerncraft can give a good overview of both in-core and memory bottlenecks and use that data to
apply performance models.
For a detailed documentation see publications in `<doc/>`_.
.. image:: https://github.com/RRZE-HPC/kerncraft/actions/workflows/test-n-publish.yml/badge.svg
:target: https://github.com/RRZE-HPC/kerncraft/actions/workflows/test-n-publish.yml
.. image:: https://codecov.io/github/RRZE-HPC/kerncraft/coverage.svg?branch=master
:target: https://codecov.io/github/RRZE-HPC/kerncraft?branch=master
Installation
============
On most systems with python pip and setuputils installed, just run:
``pip install --user kerncraft``
for the latest release. In order to get the `Intel Achitecture Code Analyzer (IACA) <https://software.intel.com/en-us/articles/intel-architecture-code-analyzer>`_, required by the `ECM`, `ECMCPU` and `RooflineASM` performance models, read `this <https://software.intel.com/protected-download/267266/157552>`_ and run:
``iaca_get --I-accept-the-Intel-What-If-Pre-Release-License-Agreement-and-please-take-my-soul``
.. warning::
As for 2023, Intel removed the download link for any IACA version. If you have any IACA version existing on your system, you can still use it with kerncraft by putting it in ``~/.kerncraft/iaca/vX.Y`` in your home directory.
Additional requirements are:
* `likwid <https://github.com/RRZE-HPC/likwid>`_ (used in Benchmark model and by ``likwid_bench_auto``)
Usage
=====
1. Get an example kernel and machine file from the examples directory
``wget https://raw.githubusercontent.com/RRZE-HPC/kerncraft/master/examples/machine-files/SandyBridgeEP_E5-2680.yml``
``wget https://raw.githubusercontent.com/RRZE-HPC/kerncraft/master/examples/kernels/2d-5pt.c``
2. Have a look at the machine file and change it to match your targeted machine (above we downloaded a file for a Sandy Bridge EP machine)
3. Run kerncraft
``kerncraft -p ECM -m SandyBridgeEP_E5-2680.yml 2d-5pt.c -D N 10000 -D M 10000``
add `-vv` for more information on the kernel and ECM model analysis.
Citations
=========
When using Kerncraft for your work, please consider citing the following publication:
`Kerncraft: A Tool for Analytic Performance Modeling of Loop Kernels <https://dx.doi.org/10.1007/978-3-319-56702-0_1>`_ (`preprint <https://arxiv.org/abs/1702.04653>`_)
::
J. Hammer, J. Eitzinger, G. Hager, and G. Wellein: Kerncraft: A Tool for Analytic Performance Modeling of Loop Kernels. In: Tools for High Performance Computing 2016, ISBN 978-3-319-56702-0, 1-22 (2017). Proceedings of IPTW 2016, the 10th International Parallel Tools Workshop, October 4-5, 2016, Stuttgart, Germany. Springer, Cham. DOI: 10.1007/978-3-319-56702-0_1, Preprint: arXiv:1702.04653``
Credits
=======
| Implementation: Julian Hammer;
| ECM Model (theory): Georg Hager, Holger Stengel, Jan Treibig;
| LC generalization: Julian Hammer
License
=======
AGPLv3
Raw data
{
"_id": null,
"home_page": "https://github.com/RRZE-HPC/kerncraft",
"name": "kerncraft",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "hpc performance benchmark analysis",
"author": "Julian Hammer",
"author_email": "julian.hammer@fau.de",
"download_url": "https://files.pythonhosted.org/packages/50/8e/2c04c4fcc5ee905755181a317e52946f30ad707cda014cca47a0bac8d816/kerncraft-0.8.16.tar.gz",
"platform": null,
"description": ".. image:: https://github.com/RRZE-HPC/kerncraft/blob/master/doc/logo/logo-lightbg.svg\n\nkerncraft\n=========\n\nLoop Kernel Analysis and Performance Modeling Toolkit\n\nThis tool allows automatic analysis of loop kernels using the `Execution Cache Memory (ECM) model <https://hpc.fau.de/research/ecm/>`_,\nthe Roofline model and actual benchmarks. kerncraft provides a framework to investigate the\ndata reuse and cache requirements by static code analysis. In combination with the Intel IACA tool\nkerncraft can give a good overview of both in-core and memory bottlenecks and use that data to\napply performance models.\n\nFor a detailed documentation see publications in `<doc/>`_.\n\n.. image:: https://github.com/RRZE-HPC/kerncraft/actions/workflows/test-n-publish.yml/badge.svg\n :target: https://github.com/RRZE-HPC/kerncraft/actions/workflows/test-n-publish.yml\n\n.. image:: https://codecov.io/github/RRZE-HPC/kerncraft/coverage.svg?branch=master\n :target: https://codecov.io/github/RRZE-HPC/kerncraft?branch=master\n\nInstallation\n============\n\nOn most systems with python pip and setuputils installed, just run:\n\n``pip install --user kerncraft``\n\nfor the latest release. In order to get the `Intel Achitecture Code Analyzer (IACA) <https://software.intel.com/en-us/articles/intel-architecture-code-analyzer>`_, required by the `ECM`, `ECMCPU` and `RooflineASM` performance models, read `this <https://software.intel.com/protected-download/267266/157552>`_ and run:\n\n``iaca_get --I-accept-the-Intel-What-If-Pre-Release-License-Agreement-and-please-take-my-soul``\n\n.. warning::\n As for 2023, Intel removed the download link for any IACA version. If you have any IACA version existing on your system, you can still use it with kerncraft by putting it in ``~/.kerncraft/iaca/vX.Y`` in your home directory.\n\nAdditional requirements are:\n * `likwid <https://github.com/RRZE-HPC/likwid>`_ (used in Benchmark model and by ``likwid_bench_auto``)\n\nUsage\n=====\n\n1. Get an example kernel and machine file from the examples directory\n\n``wget https://raw.githubusercontent.com/RRZE-HPC/kerncraft/master/examples/machine-files/SandyBridgeEP_E5-2680.yml``\n\n``wget https://raw.githubusercontent.com/RRZE-HPC/kerncraft/master/examples/kernels/2d-5pt.c``\n\n2. Have a look at the machine file and change it to match your targeted machine (above we downloaded a file for a Sandy Bridge EP machine)\n\n3. Run kerncraft\n\n``kerncraft -p ECM -m SandyBridgeEP_E5-2680.yml 2d-5pt.c -D N 10000 -D M 10000``\nadd `-vv` for more information on the kernel and ECM model analysis.\n\nCitations\n=========\n\nWhen using Kerncraft for your work, please consider citing the following publication:\n\n`Kerncraft: A Tool for Analytic Performance Modeling of Loop Kernels <https://dx.doi.org/10.1007/978-3-319-56702-0_1>`_ (`preprint <https://arxiv.org/abs/1702.04653>`_)\n\n::\n\n J. Hammer, J. Eitzinger, G. Hager, and G. Wellein: Kerncraft: A Tool for Analytic Performance Modeling of Loop Kernels. In: Tools for High Performance Computing 2016, ISBN 978-3-319-56702-0, 1-22 (2017). Proceedings of IPTW 2016, the 10th International Parallel Tools Workshop, October 4-5, 2016, Stuttgart, Germany. Springer, Cham. DOI: 10.1007/978-3-319-56702-0_1, Preprint: arXiv:1702.04653``\n\n\nCredits\n=======\n\n| Implementation: Julian Hammer;\n| ECM Model (theory): Georg Hager, Holger Stengel, Jan Treibig;\n| LC generalization: Julian Hammer\n\nLicense\n=======\nAGPLv3\n",
"bugtrack_url": null,
"license": "AGPLv3",
"summary": "Loop Kernel Analysis and Performance Modeling Toolkit",
"version": "0.8.16",
"project_urls": {
"Homepage": "https://github.com/RRZE-HPC/kerncraft"
},
"split_keywords": [
"hpc",
"performance",
"benchmark",
"analysis"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f8e6cf40177c33d75748f3e61f2462c25074253daac1ed6f456aa7df512ad583",
"md5": "2a93944a7d01e6668acbb4349c57773b",
"sha256": "8cc5551d60513181d6cb09df148799db42c9225dedb94e5e812ee8951dc1de99"
},
"downloads": -1,
"filename": "kerncraft-0.8.16-py3-none-any.whl",
"has_sig": false,
"md5_digest": "2a93944a7d01e6668acbb4349c57773b",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 125839,
"upload_time": "2024-09-04T10:51:04",
"upload_time_iso_8601": "2024-09-04T10:51:04.448866Z",
"url": "https://files.pythonhosted.org/packages/f8/e6/cf40177c33d75748f3e61f2462c25074253daac1ed6f456aa7df512ad583/kerncraft-0.8.16-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "508e2c04c4fcc5ee905755181a317e52946f30ad707cda014cca47a0bac8d816",
"md5": "fa613ea50f16150b42b102bc69c6628a",
"sha256": "70db01727993eb3c1328b7ae1b136c6f60eb41c8deac75129bc8c2f7e60edcd4"
},
"downloads": -1,
"filename": "kerncraft-0.8.16.tar.gz",
"has_sig": false,
"md5_digest": "fa613ea50f16150b42b102bc69c6628a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 672465,
"upload_time": "2024-09-04T10:51:06",
"upload_time_iso_8601": "2024-09-04T10:51:06.552942Z",
"url": "https://files.pythonhosted.org/packages/50/8e/2c04c4fcc5ee905755181a317e52946f30ad707cda014cca47a0bac8d816/kerncraft-0.8.16.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-09-04 10:51:06",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "RRZE-HPC",
"github_project": "kerncraft",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"landscape": true,
"tox": true,
"lcname": "kerncraft"
}
JSON
