{"id":1316,"date":"2023-05-28T12:16:11","date_gmt":"2023-05-28T11:16:11","guid":{"rendered":"https:\/\/blog.thomarite.uk\/?p=1316"},"modified":"2023-05-28T12:16:32","modified_gmt":"2023-05-28T11:16:32","slug":"fat-tree-drangon-fly-openai-infra","status":"publish","type":"post","link":"https:\/\/blog.thomarite.uk\/index.php\/2023\/05\/28\/fat-tree-drangon-fly-openai-infra\/","title":{"rendered":"Fat Tree &#8211; Drangonfly &#8211; OpenAI infra"},"content":{"rendered":"\n<p>I haven&#8217;t played much with ChatGPT but my first question was &#8220;how is the network infrastructure for building something like ChatGPT?&#8221; or similar. Obviously I didnt have the answer I was looking for and I think i think ask properly neither. <\/p>\n\n\n\n<p>Today, I came to this <a href=\"https:\/\/www.youtube.com\/watch?v=Rk3nTUfRZmo\">video<\/a> and at 3:30 starts something very interesting as this is an official video as says the OpenAI cluster built in 2020 for ChatGTP was actullay based on 285k AMD CPU &#8220;infinibad&#8221; plus 10k V100 GPU &#8220;infiniband&#8221; connected. They dont mention more lower level details but looks like two separated networks? And I have seen in several other pages\/videos, M$ is hardcode in infiniband.<\/p>\n\n\n\n<p>Then regarding the infiniband architectures, it seems the most common are &#8220;fat-tree&#8221; and &#8220;dragon-fly&#8221;. This <a href=\"https:\/\/www.youtube.com\/watch?v=cLSn7Q0QXG4\">video<\/a> is quite good although I have to watch it again (or more) to fully understand.<\/p>\n\n\n\n<p>These <a href=\"https:\/\/packetpushers.net\/demystifying-dcn-topologies-clos-fat-trees-part2\/\">blog<\/a>, <a href=\"https:\/\/www.cs.cornell.edu\/courses\/cs5413\/2014fa\/lectures\/08-fattree.pdf\">pdf<\/a> and <a href=\"https:\/\/en.wikipedia.org\/wiki\/Fat_tree\">wikipedia<\/a> (high level) are good for learning about &#8220;Fat-Tree&#8221;.<\/p>\n\n\n\n<p>Although most info I found is &#8220;old&#8221;, these technologies are not old. <a href=\"https:\/\/en.wikipedia.org\/wiki\/Frontier_(supercomputer)\">Frontier<\/a> and looks like most of supercomputers use it.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>I haven&#8217;t played much with ChatGPT but my first question was &#8220;how is the network infrastructure for building something like ChatGPT?&#8221; or similar. Obviously I didnt have the answer I was looking for and I think i think ask properly neither. Today, I came to this video and at 3:30 starts something very interesting as &hellip; <a href=\"https:\/\/blog.thomarite.uk\/index.php\/2023\/05\/28\/fat-tree-drangon-fly-openai-infra\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Fat Tree &#8211; Drangonfly &#8211; OpenAI infra&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[32,2],"tags":[],"class_list":["post-1316","post","type-post","status-publish","format-standard","hentry","category-cpu","category-networks"],"_links":{"self":[{"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/posts\/1316","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/comments?post=1316"}],"version-history":[{"count":3,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/posts\/1316\/revisions"}],"predecessor-version":[{"id":1319,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/posts\/1316\/revisions\/1319"}],"wp:attachment":[{"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/media?parent=1316"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/categories?post=1316"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.thomarite.uk\/index.php\/wp-json\/wp\/v2\/tags?post=1316"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}