{"id":5485,"date":"2024-03-14T09:22:38","date_gmt":"2024-03-14T01:22:38","guid":{"rendered":"https:\/\/www.1ai.net\/?p=5485"},"modified":"2024-03-14T09:22:38","modified_gmt":"2024-03-14T01:22:38","slug":"meta-%e6%96%b0%e5%bb%ba%e4%b8%a4%e5%ba%a7%e6%95%b0%e6%8d%ae%e4%b8%ad%e5%bf%83%e9%9b%86%e7%be%a4%ef%bc%9a%e5%86%85%e5%90%ab%e8%b6%85-4-9-%e4%b8%87%e5%9d%97%e8%8b%b1%e4%bc%9f%e8%be%be-h100-gpu%ef%bc%8c","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/5485.html","title":{"rendered":"Meta builds two new data center clusters: containing more than 49,000 NVIDIA H100 GPUs, dedicated to training Llama3"},"content":{"rendered":"

Meta<\/a> \u516c\u53f8\u5f53\u5730\u65f6\u95f4 12 \u65e5\u901a\u8fc7\u5b98\u65b9\u65b0\u95fb\u7a3f\u516c\u5e03\u4e86\u4e24\u5ea7\u65b0\u7684\u6570\u636e\u4e2d\u5fc3\u96c6\u7fa4\uff0c\u8be5\u516c\u53f8\u6b63\u5e0c\u671b\u901a\u8fc7\u82f1\u4f1f\u8fbe<\/a>\u7684 GPU<\/a>\uff0c\u5728\u4ee5 AI \u4e3a\u91cd\u70b9\u7684\u5f00\u53d1\u4e2d\u8131\u9896\u800c\u51fa\u3002<\/p>\n

\u636e\u6089\uff0c\u8fd9\u4e24\u5ea7\u6570\u636e\u4e2d\u5fc3\u7684\u552f\u4e00\u76ee\u7684\uff0c\u662f\u5728\u6d88\u8d39\u8005\u7279\u5b9a\u5e94\u7528\u9886\u57df\uff08IT\u4e4b\u5bb6\u6ce8\uff1a\u5305\u542b\u58f0\u97f3\u6216\u56fe\u50cf\u8bc6\u522b\uff09\u4e2d\u8fdb\u884c AI \u7814\u7a76\u548c\u5927\u8bed\u8a00\u6a21\u578b\u7684\u5f00\u53d1\uff0c\u6bcf\u4e2a\u96c6\u7fa4\u90fd\u5305\u542b\u4e86 24576 \u5757\u82f1\u4f1f\u8fbe H100 AI GPU\uff0c\u5c06\u7528\u4e8e\u81ea\u5bb6\u5927\u8bed\u8a00\u6a21\u578b Llama 3<\/a> \u7684\u8bad\u7ec3<\/span>\u3002<\/p>\n

\"aa432ce2-e4a5-4cf3-aa13-11750c8ef6a5\"<\/p>\n

\u4e24\u5ea7\u65b0\u5efa\u7684\u6570\u636e\u4e2d\u5fc3\u96c6\u7fa4\u90fd\u5177\u6709 400Gbps \u4e92\u8054\u529f\u80fd\uff0c\u5176\u4e2d\u4e00\u4e2a\u96c6\u7fa4\u91c7\u7528\u4e86 Meta \u57fa\u4e8e Arista 7800 \u81ea\u4e3b\u5f00\u53d1\u7684 Fabric \u89e3\u51b3\u65b9\u6848\uff0c\u800c\u53e6\u4e00\u4e2a\u96c6\u7fa4\u5219\u91c7\u7528\u4e86\u82f1\u4f1f\u8fbe\u7684 Quantum2 InfiniBand Fabric\uff0c\u4ee5\u786e\u4fdd\u65e0\u7f1d\u4e92\u8fde\u4f53\u9a8c\u3002<\/p>\n

\u6b64\u5916\uff0c\u96c6\u7fa4\u57fa\u4e8e Meta \u81ea\u5bb6\u7684\u5f00\u653e\u5f0f GPU Grand Teton AI \u5e73\u53f0\uff0c\u53ef\u901a\u8fc7\u63d0\u9ad8\u4e3b\u673a\u5230 GPU \u7684\u5e26\u5bbd\u548c\u8ba1\u7b97\u80fd\u529b\uff0c\u5145\u5206\u5229\u7528\u73b0\u4ee3\u52a0\u901f\u5668\u7684\u529f\u80fd\u3002<\/p>\n

Meta \u5b98\u65b9\u8868\u793a\uff0c\u8fd9\u4e9b\u96c6\u7fa4\u7684\u9ad8\u6027\u80fd\u7f51\u7edc\u7ed3\u6784\u7684\u6548\u7387\u53ca\u5173\u952e\u5b58\u50a8\u51b3\u7b56\u3001\u914d\u5408\u6bcf\u4e2a\u96c6\u7fa4\u4e2d\u7684 H100 GPU\uff0c\u80fd\u591f\u4e3a\u66f4\u5927\u3001\u66f4\u590d\u6742\u7684\u6a21\u578b\u63d0\u4f9b\u652f\u6301\uff0c\u4e3a\u901a\u7528\u4eba\u5de5\u667a\u80fd\u4ea7\u54c1\u5f00\u53d1\u3001AI \u7814\u7a76\u7684\u8fdb\u6b65\u94fa\u8def\u3002<\/p>\n

Meta \u9996\u5e2d\u6267\u884c\u5b98\u624e\u514b\u4f2f\u683c\u5ba3\u5e03\u516c\u53f8\u6b63\u5728\u5efa\u8bbe\u5e9e\u5927\u7684\u57fa\u7840\u8bbe\u65bd\u3002\u201c\u9884\u4f30\u5230\u4eca\u5e74\u5e74\u5e95\uff0c\u6211\u4eec\u5c06\u62e5\u6709\u5927\u7ea6 35 \u4e07\u7247\u82f1\u4f1f\u8fbe H100 \u52a0\u901f\u5361\uff0c\u5982\u679c\u7b97\u4e0a\u5176\u5b83 GPU \u7684\u8bdd\uff0c\u5176\u8ba1\u7b97\u80fd\u529b\u76f8\u5f53\u4e8e 60 \u4e07\u7247 H100\u3002\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

Meta, which announced two new data center clusters via an official press release on 12 local time, is looking to differentiate itself in AI-focused development with NVIDIA's GPUs. The sole purpose of the two data centers is said to be AI research and development of large language models in consumer-specific applications (IT House note: including sound or image recognition), and each cluster contains 24,576 NVIDIA H100 AI GPUs, which will be used for the training of its own large language model, Llama 3. Both new data center clusters have 400Gbps interconnectivity, with one cluster utilizing Meta's own Arista 7800-based Fabr<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[146],"tags":[415,1671,297,239],"collection":[],"class_list":["post-5485","post","type-post","status-publish","format-standard","hentry","category-news","tag-gpu","tag-llama-3","tag-meta","tag-239"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/5485","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=5485"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/5485\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=5485"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=5485"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=5485"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=5485"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}