{"id":28126,"date":"2017-09-11T09:58:57","date_gmt":"2017-09-11T07:58:57","guid":{"rendered":"http:\/\/e3mag.com\/?p=28126"},"modified":"2022-05-01T08:58:03","modified_gmt":"2022-05-01T06:58:03","slug":"120-teraflops-darchitecture","status":"publish","type":"post","link":"https:\/\/e3mag.com\/fr\/120-teraflops-darchitecture\/","title":{"rendered":"Architecture \u00e0 120 t\u00e9raflops"},"content":{"rendered":"<p>Parmi les instituts de recherche figuraient le Centre allemand de recherche en intelligence artificielle (DFKI) et l'Institut Max Planck pour les syst\u00e8mes intelligents.<\/p>\n<p>La pr\u00e9sentation a eu lieu dans le cadre de la Computer Vision and Pattern Recognition Conference \u00e0 Honolulu fin juillet et a \u00e9t\u00e9 l'un des points forts de l'\u00e9v\u00e9nement pour les plus de 150 visiteurs.<\/p>\n<p>Les GPU Tesla V100 sont bas\u00e9s sur l'architecture NVIDIA-Volta, que Huang a pr\u00e9sent\u00e9e en mai lors de la GPU Technology Conference (GTC) dans la Silicon Valley.<\/p>\n<p>Volta a \u00e9t\u00e9 con\u00e7u pour permettre des avanc\u00e9es massives dans les domaines de l'intelligence artificielle et du calcul haute performance. L'architecture est dot\u00e9e de 21 milliards de transistors et fournit une puissance d'apprentissage en profondeur de 120 t\u00e9raflops.<\/p>\n<p><a href=\"https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000.jpg\"><img loading=\"lazy\" decoding=\"async\" title=\"architecture de 120 teraflops\" class=\"aligncenter size-full wp-image-28128\" src=\"https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000.jpg\" alt=\"Nvidia, Tesla V1000\" width=\"800\" height=\"380\" srcset=\"https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000.jpg 800w, https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000-768x365.jpg 768w, https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000-100x48.jpg 100w, https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000-480x228.jpg 480w, https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000-640x304.jpg 640w, https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/Tesla-V1000-720x342.jpg 720w\" sizes=\"auto, (max-width: 800px) 100vw, 800px\" \/><\/a><\/p>","protected":false},"excerpt":{"rendered":"<p>Jensen Huang, CEO de NVIDIA, a pr\u00e9sent\u00e9 et remis dans la foul\u00e9e les premiers GPU Tesla V100 aux 15 meilleurs centres de recherche en IA de son programme NVIDIA-AI-Labs (NVAIL). <\/p>","protected":false},"author":38,"featured_media":28127,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"pmpro_default_level":"","footnotes":""},"categories":[2,16508],"tags":[625,11252,13893],"coauthors":[19920],"class_list":["post-28126","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-sap-nachrichten","category-mag-1709","tag-ai","tag-dfki","tag-nvidia","pmpro-has-access"],"acf":[],"featured_image_urls_v2":{"full":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"thumbnail":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-150x150.jpg",150,150,true],"medium":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",400,199,false],"medium_large":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-768x382.jpg",768,382,true],"large":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"image-100":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-100x50.jpg",100,50,true],"image-480":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-480x239.jpg",480,239,true],"image-640":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-640x319.jpg",640,319,true],"image-720":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-720x359.jpg",720,359,true],"image-960":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-960x478.jpg",960,478,true],"image-1168":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"image-1440":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"image-1920":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"1536x1536":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"2048x2048":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"trp-custom-language-flag":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",18,9,false],"bricks_large_16x9":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"bricks_large":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"bricks_large_square":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",1000,498,false],"bricks_medium":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",600,299,false],"bricks_medium_square":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112.jpg",600,299,false],"profile_24":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-24x24.jpg",24,24,true],"profile_48":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-48x48.jpg",48,48,true],"profile_96":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-96x96.jpg",96,96,true],"profile_150":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-150x150.jpg",150,150,true],"profile_300":["https:\/\/e3mag.com\/wp-content\/uploads\/2017\/09\/shutterstock_697380112-300x300.jpg",300,300,true]},"post_excerpt_stackable_v2":"<p>NVIDIA-CEO Jensen Huang hat den 15 Top-KI-Forschungseinrichtungen seines NVIDIA-AI-Labs-Programms (NVAIL) die ersten Tesla-V100-GPUs pr\u00e4sentiert und im Anschluss an sie \u00fcberreicht. <\/p>\n","category_list_v2":"<a href=\"https:\/\/e3mag.com\/fr\/categorie\/sap-nouvelles\/\" rel=\"category tag\">Community Nachrichten<\/a>, <a href=\"https:\/\/e3mag.com\/fr\/categorie\/mag-1709\/\" rel=\"category tag\">MAG 17-09<\/a>","author_info_v2":{"name":"Magazine E-3","url":"https:\/\/e3mag.com\/fr\/author\/e-3-magazin\/"},"comments_num_v2":"0 commentaire","_links":{"self":[{"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/posts\/28126","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/users\/38"}],"replies":[{"embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/comments?post=28126"}],"version-history":[{"count":0,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/posts\/28126\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/media\/28127"}],"wp:attachment":[{"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/media?parent=28126"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/categories?post=28126"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/tags?post=28126"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/e3mag.com\/fr\/wp-json\/wp\/v2\/coauthors?post=28126"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}