Compyle allows users to execute a restricted subset of Python on a variety of HPC platforms. It is an embedded domain-specific language (eDSL) for parallel computing. It currently supports multi-core execution using Cython, and OpenCL and CUDA for GPU devices. Users write code in a restricted subset of Python that is automatically transpiled to high-performance Cython or C. Compyle also provides a few very general purpose and useful parallel algorithms that allow users to write code once and have them run on a variety of HPC platforms. In this article, we show how to implement a simple two-dimensional molecular dynamics (MD) simulation package in pure Python using Compyle. The result is a fully parallel program that is relatively easy to implement and solves a non-trivial problem. The code transparently executes on multi-core CPUs and GPGPUs allowing simulations with millions of particles. A 3D MD code is also provided and compares very favorably with a well known, open source, molecular dynamics package.<\/p>\n","protected":false},"excerpt":{"rendered":"
Compyle allows users to execute a restricted subset of Python on a variety of HPC platforms. It is an embedded domain-specific language (eDSL) for parallel computing. It currently supports multi-core execution using Cython, and OpenCL and CUDA for GPU devices. Users write code in a restricted subset of Python that is automatically transpiled to high-performance […]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[11,89,90,3],"tags":[215,1782,14,1682,20,1793,176,513,1931],"class_list":["post-22270","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-nvidia-cuda","category-opencl","category-paper","tag-code-generation","tag-computer-science","tag-cuda","tag-hpc","tag-nvidia","tag-opencl","tag-package","tag-python","tag-tesla-p100"],"views":2163,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/22270","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=22270"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/22270\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=22270"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=22270"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=22270"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}