Deep learning has achieved remarkable success across various domains, ranging from computer vision to healthcare. General-Purpose Graphics Processing Unit (GPGPU) is one of the major driving forces behind this revolution. GPGPUs offer massive parallel computational power, enabling the training and deployment of large-scale neural networks within practical time and resource constraints. Their programmability also enables adaptability to emerging network architectures. However, entering the post-Moore\u2019s Law era, the scaling of computational power offered by GPGPUs struggles to meet the demands of novel neural networks. On the other hand, existing GPGPUs face under-utilization challenges despite the computation power shortage. This dissertation addresses the computation power shortage by improving the utilization of GPGPUs when running deep learning workloads. It presents a kernel-centric optimization approach with a focus on mapping neural networks to a more efficient set of kernels (parallel functions executed on GPGPUs) that ensures better utilization. This involves optimizations from multiple levels: algorithm level aiming to leverage more hardware-friendly formulations, operator level to harness on-chip high bandwidth on GPGPUs, and kernel implementation level that maximizes the utilization of computational resources.<\/p>\n","protected":false},"excerpt":{"rendered":"
Deep learning has achieved remarkable success across various domains, ranging from computer vision to healthcare. General-Purpose Graphics Processing Unit (GPGPU) is one of the major driving forces behind this revolution. GPGPUs offer massive parallel computational power, enabling the training and deployment of large-scale neural networks within practical time and resource constraints. Their programmability also enables […]<\/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":[11,73,89,3],"tags":[1782,1791,14,1673,34,20,2066,2115,390],"class_list":["post-29226","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-computer-vision","category-nvidia-cuda","category-paper","tag-computer-science","tag-computer-vision","tag-cuda","tag-deep-learning","tag-neural-networks","tag-nvidia","tag-nvidia-a100","tag-nvidia-v100","tag-thesis"],"views":2048,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/29226","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=29226"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/29226\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=29226"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=29226"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=29226"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}