{"id":15623,"date":"2023-05-06T06:28:50","date_gmt":"2023-05-05T21:28:50","guid":{"rendered":"https:\/\/wp.study3.biz\/?p=15623"},"modified":"2025-12-13T16:52:11","modified_gmt":"2025-12-13T07:52:11","slug":"amd-epyc-7742-64-core-processor-x2-512gb-ubuntu-20-04-6-lts-rtx-a5000-24gb-cuda-11-4-samples-nobody-benchmark-%e5%8d%98%e7%b2%be%e5%ba%a6-%e3%81%a7%e5%8b%95%e4%bd%9c%e3%81%95%e3%81%9b%e3%81%a6","status":"publish","type":"post","link":"https:\/\/wp.study3.biz\/?p=15623","title":{"rendered":"\u7b2c2\u4e16\u4ee3 AMD EPYC 7742 64-Core Processor x2 512GB  Ubuntu 20.04.6 LTS RTX A5000-24GB CUDA 11.4 Samples .\/nobody -benchmark \u5358\u7cbe\u5ea6 \u3067\u52d5\u4f5c\u3055\u305b\u3066\u307f\u305f 14061.221GFLOP\/s"},"content":{"rendered":"<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/wp.study3.biz\/wp-content\/uploads\/2021\/08\/Screenshot-from-2021-08-28-10-39-56.png\" alt=\"\" width=\"1718\" height=\"878\" class=\"alignnone size-full wp-image-8630\" \/><br \/>\nchibi@2004:~\/NVIDIA_CUDA-11.4_Samples\/5_Simulations\/nbody$ .\/nbody -benchmark -numbodies=256000 -device=0<br \/>\nRun &#8220;nbody -benchmark [-numbodies=<numBodies>]&#8221; to measure performance.<br \/>\n        -fullscreen       (run n-body simulation in fullscreen mode)<br \/>\n        -fp64             (use double precision floating point values for simulation)<br \/>\n        -hostmem          (stores simulation data in host memory)<br \/>\n        -benchmark        (run benchmark to measure performance)<br \/>\n        -numbodies=<N>    (number of bodies (>= 1) to run in simulation)<br \/>\n        -device=<d>       (where d=0,1,2&#8230;. for the CUDA device to use)<br \/>\n        -numdevices=<i>   (where i=(number of CUDA devices > 0) to use for simulation)<br \/>\n        -compare          (compares simulation results running once on the default GPU and once on the CPU)<br \/>\n        -cpu              (run n-body simulation on the CPU)<br \/>\n        -tipsy=<file.bin> (load a tipsy model file for simulation)<\/p>\n<p>NOTE: The CUDA Samples are not meant for performance measurements. Results may vary when GPU Boost is enabled.<\/p>\n<p>> Windowed mode<br \/>\n> Simulation data stored in video memory<br \/>\n> Single precision floating point simulation<br \/>\n> 1 Devices used for simulation<br \/>\ngpuDeviceInit() CUDA Device [0]: &#8220;Ampere<br \/>\n> Compute 8.6 CUDA device: [NVIDIA RTX A5000]<br \/>\nnumber of bodies = 256000<br \/>\n256000 bodies, total time for 10 iterations: 932.152 ms<br \/>\n= 703.061 billion interactions per second<br \/>\n= <strong>14061.221<\/strong> single-precision GFLOP\/s at 20 flops per interaction<br \/>\nchibi@2004:~\/NVIDIA_CUDA-11.4_Samples\/5_Simulations\/nbody$<br \/>\nRTX A5000-24GB CUDA 11.4       \u25a0\u5358\u7cbe\u5ea6\u7d50\u679c =<strong>14061.221<\/strong> single-precision GFLOP\/s at 20 flops per interaction<br \/>\nTesla V100-SXM2-16GB CUDA 10.1 \u25a0\u5358\u7cbe\u5ea6\u7d50\u679c =9033.258 single-precision GFLOP\/s at 20 flops per interaction<br \/>\n<a href=\"https:\/\/wp.study3.biz\/wp-content\/uploads\/2023\/04\/36a07e6077c7ece57f48163b0fc0835d.txt\">AMD EPYC 7742 64-Core Processor x2 512GB Ubuntu 20.04.6 LTS RTX A5000-24GB CUDA 11.4 Samples nobody benchmark \u3092\u5358\u7cbe\u5ea6 \u3067\u52d5\u4f5c\u3055\u305b\u3066\u307f\u305f 14061.221 GFLOP s\u3000deviceQuery<\/a><a href=\"https:\/\/wp.study3.biz\/wp-content\/uploads\/2023\/04\/0eefa3cc438f27523aa4684f81372102.txt\">AMD EPYC 7742 64-Core Processor x2 512GB Ubuntu 20.04.6 LTS RTX A5000-24GB CUDA 11.4 Samples nobody benchmark \u3092\u5358\u7cbe\u5ea6 \u3067\u52d5\u4f5c\u3055\u305b\u3066\u307f\u305f 14061.221 GFLOP s\u3000nvidia-smi<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>chibi@2004:~\/NVIDIA_CUDA-11.4_Samples\/5_Simulations\/nbody$ .\/nbody -benchmark -numbodies=256000 -device=0 Run  &hellip; <a href=\"https:\/\/wp.study3.biz\/?p=15623\">\u7d9a\u304d\u3092\u8aad\u3080 <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[18,17],"tags":[],"class_list":["post-15623","post","type-post","status-publish","format-standard","hentry","category-nvidia","category-ubuntu"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/15623","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=15623"}],"version-history":[{"count":6,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/15623\/revisions"}],"predecessor-version":[{"id":28850,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/15623\/revisions\/28850"}],"wp:attachment":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=15623"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=15623"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=15623"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}