This paper describes a new compiler framework for heterogeneous 3D stencil computation on GPU clusters. Our framework consists of a simple directive-based programming model and a tightly integrated source-to-source compiler. Annotated with a small number of directives, sequential stencil codes originally written in C can be automatically parallelized for large-scale GPU clusters. The most distinctive feature of the compiler is its capability to generate state-of-the-art hybrid MPI+CUDA+OpenMP code that uses concurrent CPU+GPU computing to unleash the full potential of powerful GPU clusters. At the same time, the auto-generated hybrid codes hide the overhead of various data motion by overlapping them with computation. Test results on the Titan supercomputer and the Wilkes cluster show that auto-translated codes from our compiler can achieve about 90% of the performance of highly optimized handwritten codes, for both a simple stencil benchmark and a real-world application in cardiac modeling. We thus believe that the user-friendliness and performance delivered by our domain-specific compiler framework allow computational scientists to harness the full power of GPU-accelerated supercomputing without painstaking coding effort.<\/p>\n","protected":false},"excerpt":{"rendered":"
This paper describes a new compiler framework for heterogeneous 3D stencil computation on GPU clusters. Our framework consists of a simple directive-based programming model and a tightly integrated source-to-source compiler. Annotated with a small number of directives, sequential stencil codes originally written in C can be automatically parallelized for large-scale GPU clusters. The most distinctive […]<\/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":[11,89,3],"tags":[215,955,1782,14,106,452,242,20,252,1390],"class_list":["post-17299","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-nvidia-cuda","category-paper","tag-code-generation","tag-compilers","tag-computer-science","tag-cuda","tag-gpu-cluster","tag-heterogeneous-systems","tag-mpi","tag-nvidia","tag-openmp","tag-tesla-k20"],"views":2098,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/17299","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=17299"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/17299\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=17299"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=17299"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=17299"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}