{"id":70224,"date":"2019-11-20T14:24:52","date_gmt":"2019-11-20T14:24:52","guid":{"rendered":"https:\/\/\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/"},"modified":"2019-11-20T14:24:52","modified_gmt":"2019-11-20T14:24:52","slug":"nouvelle-architecture-permettant-1-petaop-s-sur-une-puce","status":"publish","type":"post","link":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/","title":{"rendered":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce"},"content":{"rendered":"<p><span class=\"tlid-translation translation\" lang=\"fr\">L\u2019impl\u00e9mentation de la nouvelle architecture pour IA, TSP (Tensor Streaming Processor), est selon la soci\u00e9t\u00e9, \u00e9galement capable d\u2019effectuer jusqu\u2019\u00e0 250 trillions d\u2019op\u00e9rations en virgule flottante par seconde (FLOPS).<\/p>\n<p>\u00abLes plus fabricants de GPU les plus en pointe ont indiqu\u00e9 \u00e0 leurs clients qu\u2019elles esp\u00e9raient pouvoir offrir une performance PetaOp \/ s dans les prochaines ann\u00e9es alors que Groq l\u2019annonce d\u00e8s aujourd\u2019hui et \u00e9tablit ainsi une nouvelle norme de performance\u00bb, d\u00e9clare Jonathan Ross, Co-fondateur et PDG de Groq. \u00ab\u00a0L&rsquo;architecture Groq pour applications IA est beaucoup plus rapide que toutes les autres solutions disponibles pour l&rsquo;inf\u00e9rence, en termes de temps de latence r\u00e9duit et d&rsquo;inf\u00e9rences par seconde. Nos interactions avec les premiers clients le confirment.\u00a0\u00bb<\/span><br \/>\n<span class=\"tlid-translation translation\" lang=\"fr\">Selon la soci\u00e9t\u00e9, l\u2019architecture offre un nouveau paradigme permettant d\u2019obtenir \u00e0 la fois une flexibilit\u00e9 de calcul et un parall\u00e9lisme massif sans surcharge de synchronisation des architectures GPU et CPU traditionnelles. Cette architecture peut prendre en charge les mod\u00e8les d&rsquo;apprentissage automatiques traditionnels et nouveaux et est actuellement utilis\u00e9e sur les sites de clients sur des syst\u00e8mes x86 et non x86.<\/span><\/p>\n<p><span class=\"tlid-translation translation\" lang=\"fr\">Inspir\u00e9e par un esprit d\u2019abord logiciel, la nouvelle architecture de traitement plus simple est con\u00e7ue sp\u00e9cifiquement pour les exigences de performance de la vision par ordinateur, de l\u2019apprentissage automatique et d\u2019autres charges de travail li\u00e9es \u00e0 l\u2019IA. La planification de l&rsquo;ex\u00e9cution se fait par logiciel, lib\u00e9rant ainsi des fontionalit\u00e9s hardware qui seraient autrement d\u00e9di\u00e9es \u00e0 l&rsquo;ex\u00e9cution d&rsquo;instructions dynamique.<\/span><\/p>\n<p><span class=\"tlid-translation translation\" lang=\"fr\">Le contr\u00f4le strict fourni par l\u2019architecture permet un traitement d\u00e9terministe particuli\u00e8rement utile pour les applications o\u00f9 la s\u00e9curit\u00e9 et la pr\u00e9cision sont primordiales. Compar\u00e9e aux architectures traditionnelles complexes bas\u00e9es sur des processeurs, des GPU et des FPGA, la puce simplifie \u00e9galement la qualification et le d\u00e9ploiement.<\/span><\/p>\n<p><span class=\"tlid-translation translation\" lang=\"fr\">\u00ab\u00a0La solution Groq est id\u00e9ale pour le traitement \u00ab\u00a0deep learning\u00a0\u00bb des inf\u00e9rences pour un large \u00e9ventail d&rsquo;applications\u00a0\u00bb, explique Dennis Abts, architecte en chef chez Groq, \u00ab\u00a0mais m\u00eame au-del\u00e0 de cette \u00e9norme opportunit\u00e9, la solution Groq est con\u00e7ue pour une large gamme de charges de travail. Ses performances coupl\u00e9es \u00e0 sa simplicit\u00e9 en font une plate-forme id\u00e9ale pour toutes les charges de travail hautes performances, intensives en donn\u00e9es ou en calcul. \u00ab\u00a0<\/span><\/p>\n<p><span class=\"tlid-translation translation\" lang=\"fr\">Pour plus d&rsquo;informations, reportez-vous au livre blanc de la soci\u00e9t\u00e9 sur la nouvelle architecture:<\/span>&nbsp; \u00ab\u00a0<a href=\"https:\/\/www.eenewseurope.com\/Learning-center\/single-chip-petaflop-processor\">Tensor Streaming Architecture Delivers Unmatched Performance for Compute-Intensive Workloads.<\/a>\u00ab\u00a0<\/p>\n<p>Lire \u00e9galement:<\/p>\n<p><a href=\"http:\/\/Le march\u00e9 des processeurs IA d\u00e9passera les 20 milliards de dollars en 2019\">Le march\u00e9 des processeurs IA d\u00e9passera les 20 milliards de dollars en 2019<\/a><\/p>\n<p><a href=\"https:\/\/groq.com\/\">Groq<\/a><\/p>\n<p><b>Related articles:<\/b><br \/>\n<a href=\"https:\/\/www.smart2zero.com\/news\/groq-says-it-will-reveal-potent-artificial-intelligence-chip-2018\">Groq says it will reveal potent artificial intelligence chip in 2018<\/a><br \/>\n<a href=\"https:\/\/www.smart2zero.com\/news\/new-nvidia-gpu-architecture-achieves-holy-grail-computer-graphics\">New Nvidia GPU architecture achieves &lsquo;Holy Grail&rsquo; of computer graphics<\/a><br \/>\n<a href=\"https:\/\/www.smart2zero.com\/news\/steep-growth-ai-chip-market-will-produce-new-winners\">Steep growth of AI chip market will produce new winners<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>L\u2019inventeur d\u2019unit\u00e9s de traitement de tenseurs Groq (Mountain View, Californie) a annonc\u00e9 ce qui serait la premi\u00e8re architecture informatique capable de performances de 1 PetaOp \/ s (1 000.000.000.000.000 (ou un quadrillion) par seconde) sur une impl\u00e9mentation dans une simple puce unique.<\/p>\n","protected":false},"author":22,"featured_media":70225,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[883],"tags":[920,896,919,905,906,908,910,917],"domains":[47],"ppma_author":[1149],"class_list":["post-70224","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technologies","tag-artificialintelligence-fr","tag-digitalsignalprocessing-fr","tag-iot-fr","tag-memory-data-storage-fr","tag-mpus-mcus-fr","tag-plds-fpgas-asics-fr","tag-sensing-conditioning-fr","tag-software-embedded-tools-fr","domains-electronique-eci"],"acf":[],"yoast_head":"<title>Nouvelle architecture permettant 1 PetaOp\/s sur une puce ...<\/title>\n<meta name=\"description\" content=\"L\u2019inventeur d\u2019unit\u00e9s de traitement de tenseurs Groq (Mountain View, Californie) a annonc\u00e9 ce qui serait la premi\u00e8re architecture informatique capable...\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Nouvelle architecture permettant 1 PetaOp\/s sur une puce\" \/>\n<meta property=\"og:description\" content=\"L\u2019inventeur d\u2019unit\u00e9s de traitement de tenseurs Groq (Mountain View, Californie) a annonc\u00e9 ce qui serait la premi\u00e8re architecture informatique capable de performances de 1 PetaOp \/ s (1 000.000.000.000.000 (ou un quadrillion) par seconde) sur une impl\u00e9mentation dans une simple puce unique.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224\/\" \/>\n<meta property=\"og:site_name\" content=\"EENewsEurope\" \/>\n<meta property=\"article:published_time\" content=\"2019-11-20T14:24:52+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.ecinews.fr\/wp-content\/uploads\/import\/default\/files\/sites\/default\/files\/images\/2019-11-15_tensor_compute_architecture_groq_ai_machine_learning.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"600\" \/>\n\t<meta property=\"og:image:height\" content=\"350\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"eeNews Europe\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"eeNews Europe\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/\"},\"author\":{\"name\":\"eeNews Europe\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/9eff4051fa9dac8230052de45e32b0f4\"},\"headline\":\"Nouvelle architecture permettant 1 PetaOp\/s sur une puce\",\"datePublished\":\"2019-11-20T14:24:52+00:00\",\"dateModified\":\"2019-11-20T14:24:52+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/\"},\"wordCount\":482,\"publisher\":{\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#organization\"},\"keywords\":[\"ArtificialIntelligence\",\"DigitalSignalProcessing\",\"IoT\",\"Memory &amp; Data Storage\",\"MPUs\/MCUs\",\"PLDs\/FPGAs\/ASICs\",\"Sensing \/ Conditioning\",\"Software &amp; Embedded tools\"],\"articleSection\":[\"Technologies\"],\"inLanguage\":\"fr-FR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/\",\"url\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/\",\"name\":\"Nouvelle architecture permettant 1 PetaOp\/s sur une puce -\",\"isPartOf\":{\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#website\"},\"datePublished\":\"2019-11-20T14:24:52+00:00\",\"dateModified\":\"2019-11-20T14:24:52+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.ecinews.fr\/fr\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Nouvelle architecture permettant 1 PetaOp\/s sur une puce\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#website\",\"url\":\"https:\/\/www.eenewseurope.com\/en\/\",\"name\":\"EENewsEurope\",\"description\":\"Just another WordPress site\",\"publisher\":{\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.eenewseurope.com\/en\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#organization\",\"name\":\"EENewsEurope\",\"url\":\"https:\/\/www.eenewseurope.com\/en\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.ecinews.fr\/wp-content\/uploads\/2022\/02\/logo-1.jpg\",\"contentUrl\":\"https:\/\/www.ecinews.fr\/wp-content\/uploads\/2022\/02\/logo-1.jpg\",\"width\":283,\"height\":113,\"caption\":\"EENewsEurope\"},\"image\":{\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/9eff4051fa9dac8230052de45e32b0f4\",\"name\":\"eeNews Europe\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/image\/fae8f0cb15861c4ae0ed4872e2c9fc22\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/5081509054e28b04ecd976976e723ce0?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/5081509054e28b04ecd976976e723ce0?s=96&d=mm&r=g\",\"caption\":\"eeNews Europe\"}}]}<\/script>","yoast_head_json":{"title":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce ...","description":"L\u2019inventeur d\u2019unit\u00e9s de traitement de tenseurs Groq (Mountain View, Californie) a annonc\u00e9 ce qui serait la premi\u00e8re architecture informatique capable...","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224\/","og_locale":"fr_FR","og_type":"article","og_title":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce","og_description":"L\u2019inventeur d\u2019unit\u00e9s de traitement de tenseurs Groq (Mountain View, Californie) a annonc\u00e9 ce qui serait la premi\u00e8re architecture informatique capable de performances de 1 PetaOp \/ s (1 000.000.000.000.000 (ou un quadrillion) par seconde) sur une impl\u00e9mentation dans une simple puce unique.","og_url":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224\/","og_site_name":"EENewsEurope","article_published_time":"2019-11-20T14:24:52+00:00","og_image":[{"width":600,"height":350,"url":"https:\/\/www.ecinews.fr\/wp-content\/uploads\/import\/default\/files\/sites\/default\/files\/images\/2019-11-15_tensor_compute_architecture_groq_ai_machine_learning.jpg","type":"image\/jpeg"}],"author":"eeNews Europe","twitter_card":"summary_large_image","twitter_misc":{"Written by":"eeNews Europe","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#article","isPartOf":{"@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/"},"author":{"name":"eeNews Europe","@id":"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/9eff4051fa9dac8230052de45e32b0f4"},"headline":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce","datePublished":"2019-11-20T14:24:52+00:00","dateModified":"2019-11-20T14:24:52+00:00","mainEntityOfPage":{"@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/"},"wordCount":482,"publisher":{"@id":"https:\/\/www.eenewseurope.com\/en\/#organization"},"keywords":["ArtificialIntelligence","DigitalSignalProcessing","IoT","Memory &amp; Data Storage","MPUs\/MCUs","PLDs\/FPGAs\/ASICs","Sensing \/ Conditioning","Software &amp; Embedded tools"],"articleSection":["Technologies"],"inLanguage":"fr-FR"},{"@type":"WebPage","@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/","url":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/","name":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce -","isPartOf":{"@id":"https:\/\/www.eenewseurope.com\/en\/#website"},"datePublished":"2019-11-20T14:24:52+00:00","dateModified":"2019-11-20T14:24:52+00:00","breadcrumb":{"@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.ecinews.fr\/fr\/nouvelle-architecture-permettant-1-petaop-s-sur-une-puce\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.ecinews.fr\/fr\/"},{"@type":"ListItem","position":2,"name":"Nouvelle architecture permettant 1 PetaOp\/s sur une puce"}]},{"@type":"WebSite","@id":"https:\/\/www.eenewseurope.com\/en\/#website","url":"https:\/\/www.eenewseurope.com\/en\/","name":"EENewsEurope","description":"Just another WordPress site","publisher":{"@id":"https:\/\/www.eenewseurope.com\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.eenewseurope.com\/en\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fr-FR"},{"@type":"Organization","@id":"https:\/\/www.eenewseurope.com\/en\/#organization","name":"EENewsEurope","url":"https:\/\/www.eenewseurope.com\/en\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/www.eenewseurope.com\/en\/#\/schema\/logo\/image\/","url":"https:\/\/www.ecinews.fr\/wp-content\/uploads\/2022\/02\/logo-1.jpg","contentUrl":"https:\/\/www.ecinews.fr\/wp-content\/uploads\/2022\/02\/logo-1.jpg","width":283,"height":113,"caption":"EENewsEurope"},"image":{"@id":"https:\/\/www.eenewseurope.com\/en\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/9eff4051fa9dac8230052de45e32b0f4","name":"eeNews Europe","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/www.eenewseurope.com\/en\/#\/schema\/person\/image\/fae8f0cb15861c4ae0ed4872e2c9fc22","url":"https:\/\/secure.gravatar.com\/avatar\/5081509054e28b04ecd976976e723ce0?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/5081509054e28b04ecd976976e723ce0?s=96&d=mm&r=g","caption":"eeNews Europe"}}]}},"authors":[{"term_id":1149,"user_id":22,"is_guest":0,"slug":"eenews-europe","display_name":"eeNews Europe","avatar_url":"https:\/\/secure.gravatar.com\/avatar\/5081509054e28b04ecd976976e723ce0?s=96&d=mm&r=g","0":null,"1":"","2":"","3":"","4":"","5":"","6":"","7":"","8":""}],"_links":{"self":[{"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224"}],"collection":[{"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/users\/22"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/comments?post=70224"}],"version-history":[{"count":0,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/posts\/70224\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/media\/70225"}],"wp:attachment":[{"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/media?parent=70224"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/categories?post=70224"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/tags?post=70224"},{"taxonomy":"domains","embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/domains?post=70224"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.ecinews.fr\/fr\/wp-json\/wp\/v2\/ppma_author?post=70224"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}