{"id":6604,"date":"2011-12-16T16:15:17","date_gmt":"2011-12-16T14:15:17","guid":{"rendered":"http:\/\/hgpu.org\/?p=6604"},"modified":"2011-12-16T16:15:17","modified_gmt":"2011-12-16T14:15:17","slug":"a-parallel-gpu-version-of-the-traveling-salesman-problem","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=6604","title":{"rendered":"A Parallel GPU Version of the Traveling Salesman Problem"},"content":{"rendered":"<p>This paper describes and evaluates an implementation of iterative hill climbing with random restart for determining high-quality solutions to the traveling salesman problem. With 100,000 restarts, this algorithm finds the optimal solution for four out of five 100-city TSPLIB inputs and yields a tour that is only 0.07% longer than the optimum on the fifth input. The presented implementation is highly parallel and optimized for GPU-based execution. Running on a single GPU, it evaluates over 20 billion tour modifications per second. It takes 32 CPUs with 8 cores each (256 cores total) to match this performance.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>This paper describes and evaluates an implementation of iterative hill climbing with random restart for determining high-quality solutions to the traveling salesman problem. With 100,000 restarts, this algorithm finds the optimal solution for four out of five 100-city TSPLIB inputs and yields a tour that is only 0.07% longer than the optimum on the fifth [&hellip;]<\/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":[1787,1782,14,20,298,176,378],"class_list":["post-6604","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-nvidia","tag-optimization","tag-package","tag-tesla-c2050"],"views":3000,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/6604","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=6604"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/6604\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6604"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6604"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6604"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}