Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize the computational performance of deep learning frameworks. However, the same cannot be said for I\/O performance. As deep learning algorithms rely on big-data volume and variety to effectively train neural networks accurately, I\/O is a significant bottleneck on large-scale distributed deep learning training. This study aims to provide a detailed investigation of the I\/O behavior of various scientific deep learning workloads running on the Theta supercomputer at Argonne Leadership Computing Facility. In this paper, we present DLIO, a novel representative benchmark suite built based on the I\/O profiling of the selected workloads. DLIO can be utilized to accurately emulate the I\/O behavior of modern scientific deep learning applications. Using DLIO, application developers and system software solution architects can identify potential I\/O bottlenecks in their applications and guide optimizations to boost the I\/O performance leading to lower training times by up to 6.7x.<\/p>\n","protected":false},"excerpt":{"rendered":"
Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize […]<\/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":[11,3],"tags":[451,1782,1673,242,34,176,1909],"class_list":["post-25107","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-paper","tag-benchmarking","tag-computer-science","tag-deep-learning","tag-mpi","tag-neural-networks","tag-package","tag-tensorflow"],"views":2027,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/25107","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=25107"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/25107\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=25107"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=25107"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=25107"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}