{"id":11276,"date":"2014-01-23T23:08:34","date_gmt":"2014-01-23T21:08:34","guid":{"rendered":"http:\/\/hgpu.org\/?p=11276"},"modified":"2014-01-23T23:08:34","modified_gmt":"2014-01-23T21:08:34","slug":"clpeak-peak-performance-of-your-opencl-device","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=11276","title":{"rendered":"clpeak &#8211; peak performance of your opencl device"},"content":{"rendered":"<p>clpeak is a benchmarking tool intended toward developers to fine-tune opencl kernels for a particular device\/class of device. It calculates bandwidth &amp; compute performance for different vector-widths of a datatype, say float, float4. Traditionally it is recommended to use scalar code and we expect opencl compiler to auto-vectorize it. But, most of the times compiler will not be able to vectorize a scalar code. A hand-written vector code is always efficient in performance critical scenarios. This tool gives an idea about internal architecture of the device and what vector-widths should be used to realize full potential. It also measures host to device transfer bandwidths and vice-versa. Transfers can be done using enqueueWriteBuffer or enqueueMapBuffer. Map can happen through pinned-memory or sometimes zero-copy. This tool can indicate a zero-copy transfer and memcpy bandwidth on zero-copied memory.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>clpeak is a benchmarking tool intended toward developers to fine-tune opencl kernels for a particular device\/class of device. It calculates bandwidth &amp; compute performance for different vector-widths of a datatype, say float, float4. Traditionally it is recommended to use scalar code and we expect opencl compiler to auto-vectorize it. But, most of the times compiler [&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":[7,451,1782,20,1793,176,67],"class_list":["post-11276","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-opencl","category-paper","tag-ati","tag-benchmarking","tag-computer-science","tag-nvidia","tag-opencl","tag-package","tag-performance"],"views":4694,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/11276","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=11276"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/11276\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=11276"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=11276"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=11276"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}