We present a performance model-driven framework for automated performance tuning (autotuning) of sparse matrix-vector multiply (SpMV) on systems accelerated by graphics processing units (GPU). Our study consists of two parts. First, we describe several carefully hand-tuned SpMV implementations for GPUs, identifying key GPU-specific performance limitations, enhancements, and tuning opportunities. These implementations, which include variants on classical blocked compressed sparse row (BCSR) and blocked ELLPACK (BELLPACK) storage formats, match or exceed state-of-the-art implementations. For instance, our best BELLPACK implementation achieves up to 29.0 Gflop\/s in single-precision and 15.7 Gflop\/s in double-precision on the NVIDIA T10P multiprocessor (C1060), enhancing prior state-of-the-art unblocked implementations (Bell and Garland, 2009) by up to 1.8? and 1.5? for single-and double-precision respectively. However, achieving this level of performance requires input matrix-dependent parameter tuning. Thus, in the second part of this study, we develop a performance model that can guide tuning. Like prior autotuning models for CPUs (e.g., Im, Yelick, and Vuduc, 2004), this model requires offline measurements and run-time estimation, but more directly models the structure of multithreaded vector processors like GPUs. We show that our model can identify the implementations that achieve within 15% of those found through exhaustive search.<\/p>\n","protected":false},"excerpt":{"rendered":"
We present a performance model-driven framework for automated performance tuning (autotuning) of sparse matrix-vector multiply (SpMV) on systems accelerated by graphics processing units (GPU). Our study consists of two parts. First, we describe several carefully hand-tuned SpMV implementations for GPUs, identifying key GPU-specific performance limitations, enhancements, and tuning opportunities. These implementations, which include variants on […]<\/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":[36,11,89,3],"tags":[1787,1782,14,37,20,67,421,199,202,851],"class_list":["post-2729","post","type-post","status-publish","format-standard","hentry","category-algorithms","category-computer-science","category-nvidia-cuda","category-paper","tag-algorithms","tag-computer-science","tag-cuda","tag-linear-algebra","tag-nvidia","tag-performance","tag-sparse-matrix","tag-tesla-c1060","tag-tesla-c870","tag-tesla-t10p"],"views":2915,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/2729","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=2729"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/2729\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2729"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2729"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2729"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}