To cope with the complexity of programming GPU accelerators for medical imaging computations, we developed a framework to describe image processing kernels in a domainspecific language, which is embedded into C++. The description uses decoupled access\/execute metadata, which allow the programmer to specify both execution constraints and memory access patterns of kernels. A source-to-source compiler translates this high-level description into low-level CUDA and OpenCL code with automatic support for boundary handling and filter masks. Taking the annotated metadata and the characteristics of the parallel GPU execution model into account, two-layered parallel implementations-utilizing SPMD and MPMD parallelism are generated. An abstract hardware model of graphics card architectures allows to model GPUs of multiple vendors like AMD and NVIDIA, and to generate device-specific code for multiple targets. It is shown that the generated code is faster than manual implementations and those relying on hardware support for boundary handling. Implementations from RapidMind, a commercial framework for GPU programming, are outperformed and similar results achieved compared to the GPU backend of the widely used image processing library OpenCV.<\/p>\n","protected":false},"excerpt":{"rendered":"
To cope with the complexity of programming GPU accelerators for medical imaging computations, we developed a framework to describe image processing kernels in a domainspecific language, which is embedded into C++. The description uses decoupled access\/execute metadata, which allow the programmer to specify both execution constraints and memory access patterns of kernels. A source-to-source compiler […]<\/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":[89,33,38,90,3],"tags":[7,455,1200,215,14,1786,1788,20,710,1793,947,378],"class_list":["post-7684","post","type-post","status-publish","format-standard","hentry","category-nvidia-cuda","category-image-processing","category-medicine","category-opencl","category-paper","tag-ati","tag-ati-radeon-hd-5870","tag-ati-radeon-hd-6970","tag-code-generation","tag-cuda","tag-image-processing","tag-medicine","tag-nvidia","tag-nvidia-quadro-fx-5800","tag-opencl","tag-rapidmind","tag-tesla-c2050"],"views":2641,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/7684","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=7684"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/7684\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7684"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7684"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7684"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}