Nowadays, it is possible to build a multi-GPU supercomputer, well suited for implementation of digital signal processing algorithms, for a few thousand dollars. However, to achieve the highest performance with this kind of architecture, the programmer has to focus on inter-processor communications, tasks synchronization. In this paper, we propose a high level programming model based on a data flow graph (DFG) allowing an efficient implementation of digital signal processing applications on a multi-GPU computer cluster. This DFG-based design flow abstracts the underlying architecture. We focus particularly on the efficient implementation of communications by automating computation-communication overlap, which can lead to significant speedups as shown in the presented benchmark. The approach is validated on three experiments: a multi-host multi-gpu benchmark, a 3D granulometry application developed for research on materials and an application for computing visual saliency maps.<\/p>\n","protected":false},"excerpt":{"rendered":"
Nowadays, it is possible to build a multi-GPU supercomputer, well suited for implementation of digital signal processing algorithms, for a few thousand dollars. However, to achieve the highest performance with this kind of architecture, the programmer has to focus on inter-processor communications, tasks synchronization. In this paper, we propose a high level programming model based […]<\/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":[36,89,33,3,41],"tags":[1787,14,106,1786,20,251,1187,1789],"class_list":["post-8468","post","type-post","status-publish","format-standard","hentry","category-algorithms","category-nvidia-cuda","category-image-processing","category-paper","category-signal-processing","tag-algorithms","tag-cuda","tag-gpu-cluster","tag-image-processing","tag-nvidia","tag-nvidia-geforce-gtx-285","tag-nvidia-geforce-gtx-460-m","tag-signal-processing"],"views":2835,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/8468","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=8468"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/8468\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8468"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8468"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8468"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}