{"id":16373,"date":"2016-08-04T00:20:46","date_gmt":"2016-08-03T21:20:46","guid":{"rendered":"http:\/\/hgpu.org\/?p=16373"},"modified":"2016-08-04T00:20:46","modified_gmt":"2016-08-03T21:20:46","slug":"trees-a-cpugpu-task-parallel-runtime-with-explicit-epoch-synchronization","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=16373","title":{"rendered":"TREES: A CPU\/GPU Task-Parallel Runtime with Explicit Epoch Synchronization"},"content":{"rendered":"<p>We have developed a task-parallel runtime system, called TREES, that is designed for high performance on CPU\/GPU platforms. On platforms with multiple CPUs, Cilk&#8217;s &quot;work-first&quot; principle underlies how task-parallel applications can achieve performance, but work-first is a poor fit for GPUs. We build upon work-first to create the &quot;work-together&quot; principle that addresses the specific strengths and weaknesses of GPUs. The work-together principle extends work-first by stating that (a) the overhead on the critical path should be paid by the entire system at once and (b) work overheads should be paid co-operatively. We have implemented the TREES runtime in OpenCL, and we experimentally evaluate TREES applications on a CPU\/GPU platform.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>We have developed a task-parallel runtime system, called TREES, that is designed for high performance on CPU\/GPU platforms. On platforms with multiple CPUs, Cilk&#8217;s &quot;work-first&quot; principle underlies how task-parallel applications can achieve performance, but work-first is a poor fit for GPUs. We build upon work-first to create the &quot;work-together&quot; principle that addresses the specific strengths [&hellip;]<\/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":[11,90,3],"tags":[1438,1197,1782,1793,852,660],"class_list":["post-16373","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-opencl","category-paper","tag-amd","tag-apu","tag-computer-science","tag-opencl","tag-operating-systems","tag-programming-languages"],"views":2475,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/16373","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=16373"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/16373\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=16373"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=16373"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=16373"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}