An efficient memory bandwidth utilization for GPU accelerators is crucial for memory bound applications. In medical imaging, the performance of many kernels is limited by the available memory bandwidth since only a few operations are performed per pixel. For such kernels only a fraction of the compute power provided by GPU accelerators can be exploited and performance is predetermined by memory bandwidth. As a remedy, this paper investigates the optimal utilization of available memory bandwidth by means of increasing in-flight memory transactions. Instead of doing this manually for different GPU accelerators, the required CUDA and OpenCL code is automatically generated from descriptions in a Domain-Specific Language (DSL) for the considered application domain. Moreover, the DSL is extended to also support global reduction operators. We show that the generated target-specific code improves bandwidth utilization for memory-bound kernels significantly. Moreover, competitive performance compared to the GPU back end of the widely used image processing library OpenCV can be achieved.<\/p>\n","protected":false},"excerpt":{"rendered":"
An efficient memory bandwidth utilization for GPU accelerators is crucial for memory bound applications. In medical imaging, the performance of many kernels is limited by the available memory bandwidth since only a few operations are performed per pixel. For such kernels only a fraction of the compute power provided by GPU accelerators can be exploited […]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"open","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":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[89,33,90,3],"tags":[215,14,1786,20,1793,298,176,378],"class_list":["post-7842","post","type-post","status-publish","format-standard","hentry","category-nvidia-cuda","category-image-processing","category-opencl","category-paper","tag-code-generation","tag-cuda","tag-image-processing","tag-nvidia","tag-opencl","tag-optimization","tag-package","tag-tesla-c2050"],"views":2435,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/7842","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=7842"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/7842\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7842"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7842"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7842"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}