{"id":508979,"date":"2023-05-24T22:34:40","date_gmt":"2023-05-24T20:34:40","guid":{"rendered":"https:\/\/www.devoteam.com\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/"},"modified":"2025-02-04T16:08:09","modified_gmt":"2025-02-04T15:08:09","slug":"botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines","status":"publish","type":"success-story","link":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/","title":{"rendered":"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\"><strong>Le Bon Coin, Expedia, Glassdoor, Deezer, The New York Times, L\u2019Or\u00e9al\u2026 toutes ces entreprises ont comme point commun de g\u00e9rer un grand volume de pages web. Afin d\u2019optimiser leur r\u00e9f\u00e9rencement dans les moteurs de recherche, elles font appel \u00e0&nbsp;<a href=\"https:\/\/www.botify.com\/\" target=\"_blank\" rel=\"noreferrer noopener\">Botify<\/a>&nbsp;: l\u2019entreprise crawle 5 milliards de pages par mois, soit plus de 2 petabytes de donn\u00e9es. Pour acc\u00e9l\u00e9rer le temps de d\u00e9veloppement et de mise en production de ses mod\u00e8les de Machine Learning, l\u2019\u00e9quipe Data Science de Botify a \u00e9labor\u00e9 en collaboration avec Devoteam Revolve une solution bas\u00e9e sur Amazon SageMaker Pipelines.<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Cet article propose un compte-rendu du talk \u00ab&nbsp;Passez \u00e0 la vitesse sup\u00e9rieure avec Amazon Sagemaker Pipelines&nbsp;\u00bb donn\u00e9 au&nbsp;<strong>Summit AWS Paris du 12 Avril 2022<\/strong>, par Yanal Wazaefi (Botify), Fabien Lallemand (Devoteam Revolve) et Jamal Aourraz (Devoteam Revolve).<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Comme l\u2019a expliqu\u00e9&nbsp;<strong>Yanal Wazaefi<\/strong>, Head of Data Science Botify, ces derni\u00e8res ann\u00e9es les moteurs de recherche rencontrent des difficult\u00e9s pour recenser l\u2019ensemble des pages Web. Le contenu a explos\u00e9 en volume, et les moteurs de recherche peinent \u00e0 interpr\u00e9ter le contenu dynamique des pages. Certains sites ont ainsi des&nbsp;<strong>dizaines de millions de pages<\/strong>, comme Le Bon Coin, Carrefour, ou Fnac Darty, et Botify propose \u00e0 ces entreprises des solutions pour optimiser techniquement leurs sites,&nbsp;<strong>faciliter l\u2019indexation<\/strong>&nbsp;et donc&nbsp;<strong>optimiser le r\u00e9f\u00e9rencement naturel<\/strong>, ou Search Engine Optimization (SEO).<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter\"><img decoding=\"async\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Devoteam_Revolve_Blog_Botify_SagemakerPipelines-4.jpg\" alt=\"\" class=\"wp-image-18166\"\/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Acc\u00e9l\u00e9rer les mises en production des mod\u00e8les<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Les \u00e9quipes de Data Scientists de Botify travaillent pour&nbsp;<strong>g\u00e9n\u00e9rer des recommandations intelligentes<\/strong>&nbsp;et optimiser le r\u00e9f\u00e9rencement en utilisant des algorithmes de Machine Learning, de Natural Language Processing, ou des statistiques avanc\u00e9es. Le cycle de projet est relativement classique : les Data Scientists explorent et prototypent dans des&nbsp;<strong>notebook Jupyter<\/strong>, ils valident ensuite leurs r\u00e9sultats avec les experts m\u00e9tiers et SEO, puis ils collaborent avec les ing\u00e9nieurs charg\u00e9s de l\u2019int\u00e9gration de ces solutions dans les workflows, qui sont bas\u00e9s sur le service&nbsp;<strong>AWS Simple Workflow<\/strong>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Ce cycle de projet pouvait prendre jusqu\u2019\u00e0 plusieurs mois et l\u2019\u00e9quipe Botify souhaitait aller plus vite, et pouvoir :<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>It\u00e9rer pour tester les nouveaux mod\u00e8les<\/li>\n\n\n\n<li>Disposer de plusieurs environnements<\/li>\n\n\n\n<li>Mettre en place une int\u00e9gration continue<\/li>\n\n\n\n<li>Monitorer les pipelines<\/li>\n\n\n\n<li>Suivre les performances des mod\u00e8les<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">L\u2019entreprise a fait appel \u00e0 l\u2019\u00e9quipe Devoteam A Cloud pour les aider \u00e0 trouver une solution \u00e0 ces probl\u00e9matiques<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/image-8-1.png\" alt=\"\" class=\"wp-image-12456\" width=\"706\" height=\"470\"\/><\/figure>\n\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><strong>Yanal Wazaefi<\/strong>, Head of Data Science Botify<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Les contraintes du Machine Learning en production<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Pour&nbsp;<strong>Fabien Lallemand<\/strong>, Devoteam Revolve, il est essentiel de clarifier ce qu\u2019on entend par \u201cmachine learning en production\u201d. De nombreuses entreprises pensent avoir du ML en production avec un cas d\u2019usage qui tourne sur le poste d\u2019un Data Scientist, mais le plus souvent beaucoup d\u2019actions, comme la r\u00e9cup\u00e9ration des features, doivent \u00eatre faites manuellement.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">On parlera ici de mise en production au sens IT du terme, et c\u2019est une phase du projet qui implique trois \u00e9quipes distinctes, Data Scientists, ing\u00e9nieurs logiciels, et Ops, avec des objectifs diff\u00e9rents et parfois contraires, et qui pourtant doivent collaborer vers un objectif commun. La difficult\u00e9 est de&nbsp;<strong>g\u00e9rer ce partage des contraintes<\/strong>&nbsp;entre les diff\u00e9rentes parties.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">C\u2019est le concept du&nbsp;<strong><a href=\"https:\/\/blog.revolve.team\/2022\/03\/10\/machine-learning-production-ml-ops-cloud\/\" target=\"_blank\" rel=\"noreferrer noopener\">ML Ops<\/a><\/strong>&nbsp;:<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img decoding=\"async\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/image-11.png\" alt=\"\" class=\"wp-image-12474\"\/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Dans le cas d\u2019un pipeline classique de Machine Learning, le principal probl\u00e8me rencontr\u00e9 lors du passage en production \u00e9tait la&nbsp;<strong>scalabilit\u00e9<\/strong>&nbsp;: ce qui fonctionnait avec un petit dataset ne passait plus avec l\u2019int\u00e9gralit\u00e9 des donn\u00e9es en production. Aujourd\u2019hui ce point est g\u00e9r\u00e9 nativement par la plupart des algorithmes, mais il reste quatre points \u00e0 adresser :<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>L\u2019int\u00e9gration<\/strong>&nbsp;dans l\u2019\u00e9cosyst\u00e8me IT : comment faire de l\u2019algorithme une application<\/li>\n\n\n\n<li><strong>La pr\u00e9paration<\/strong>&nbsp;de la donn\u00e9e : la feature va \u00eatre utilis\u00e9e au moment de l\u2019entra\u00eenement du mod\u00e8le mais aussi au moment de l\u2019inf\u00e9rence temps r\u00e9el, durant laquelle les temps de r\u00e9ponse doivent \u00eatre beaucoup plus courts.<\/li>\n\n\n\n<li>La d\u00e9tection des&nbsp;<strong>performances<\/strong>&nbsp;: pouvoir d\u00e9tecter les baisses de performance du mod\u00e8le en production qui sont dues \u00e0 des causes ext\u00e9rieures (par exemple la d\u00e9tection des visages qui ne fonctionne plus \u00e0 cause des masques Covid)<\/li>\n\n\n\n<li>La&nbsp;<strong>standardisation<\/strong>&nbsp;et la g\u00e9n\u00e9ralisation du pipeline de ML.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Sur ce dernier point, Amazon SageMaker Pipelines am\u00e8ne une standardisation et permet de traiter de la m\u00eame mani\u00e8re un algorithme supervis\u00e9 ou non supervis\u00e9, ou par exemple de traiter un cas d\u2019usage de d\u00e9tection d\u2019image. Tout sera packag\u00e9 avec la m\u00eame interface et on aura donc un pipeline de livraison homog\u00e8ne.<\/p>\n\n\n\n<p class=\"has-text-align-center wp-block-paragraph\"><em>Amazon SageMaker Pipelines am\u00e8ne une standardisation et permet de traiter de la m\u00eame mani\u00e8re un algorithme supervis\u00e9 ou non supervis\u00e9<\/em>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Par ailleurs, quand il faut r\u00e9-entra\u00eener le mod\u00e8le, on doit repasser par l\u2019ensemble des environnements (d\u00e9veloppement, pr\u00e9-production, production\u2026). Dans le cas de Botify, le d\u00e9veloppement d\u2019une solution bas\u00e9e sur Amazon SageMaker Pipelines permet maintenant de&nbsp;<strong>livrer la m\u00e9canique globale de r\u00e9-entra\u00eenement du mod\u00e8le<\/strong>, plut\u00f4t que de livrer simplement le mod\u00e8le. Cette m\u00e9thode et la livraison du pipeline permet \u00e0 Botify de disposer d\u2019une cha\u00eene unique pour&nbsp;<strong>reconstruire le mod\u00e8le pour chacun des cas d\u2019usage<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Architecture de la solution<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Quels composants AWS ont \u00e9t\u00e9 utilis\u00e9s pour r\u00e9pondre au besoin de Botify ?<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">L\u2019\u00e9co-syst\u00e8me se base sur Amazon SageMaker et plus particuli\u00e8rement sur&nbsp;<strong><a href=\"https:\/\/aws.amazon.com\/fr\/sagemaker\/\" target=\"_blank\" rel=\"noreferrer noopener\">SageMaker Pipelines<\/a><\/strong>&nbsp;et&nbsp;<strong><a href=\"https:\/\/docs.aws.amazon.com\/sagemaker\/latest\/dg\/model-registry.html\" target=\"_blank\" rel=\"noreferrer noopener\">SageMaker Model Registry<\/a><\/strong>. Ce projet, men\u00e9 en 2021, a \u00e9t\u00e9 l\u2019un des premiers en France \u00e0 tirer parti de Sagemaker Pipelines, qui \u00e0 ce moment venait tout juste de passer en&nbsp;<em>general availability<\/em>. SageMaker Pipelines est la solution ML Ops d\u2019AWS, et apporte les pratiques et process DevOps appliqu\u00e9s au Machine Learning.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img decoding=\"async\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/image-9-1.png\" alt=\"\" class=\"wp-image-12462\"\/><\/figure>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img decoding=\"async\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/image-10.png\" alt=\"\" class=\"wp-image-12468\"\/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">La solution d\u00e9velopp\u00e9e autour d\u2019Amazon SageMaker Pipelines a permis de r\u00e9pondre aux objectifs de Botify :<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Assurer la scalabilit\u00e9<\/li>\n\n\n\n<li>R\u00e9duire le temps de mise en production<\/li>\n\n\n\n<li>Donner de l\u2019autonomie aux Data Scientists<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Aujourd\u2019hui l\u2019\u00e9quipe Data Botify a tripl\u00e9 ses effectifs, et a pu ajouter plus de 100 \u00ab&nbsp;smart features&nbsp;\u00bb \u00e0 ses solutions.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Le Bon Coin, Expedia, Glassdoor, Deezer, The New York Times, L\u2019Or\u00e9al\u2026 toutes ces entreprises ont comme point commun de g\u00e9rer un grand volume de pages web. Afin d\u2019optimiser leur r\u00e9f\u00e9rencement dans les moteurs de recherche, elles font appel \u00e0&nbsp;Botify&nbsp;: l\u2019entreprise crawle 5 milliards de pages par mois, soit plus de 2 petabytes de donn\u00e9es. Pour [&hellip;]<\/p>\n","protected":false},"featured_media":509086,"template":"","categories":[1358,1363],"tags":[585,212],"industry":[],"class_list":["post-508979","success-story","type-success-story","status-publish","has-post-thumbnail","hentry","category-aws-fr","category-data-fr","tag-aws","tag-data"],"acf":[],"cards":"\n\t<div class=\"single-post-card\">\n\n\t\t<figure class=\"wp-block-post-featured-image\"><a href=\"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/\" target=\"_self\" ><img width=\"960\" height=\"540\" src=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png\" class=\"attachment-post-thumbnail size-post-thumbnail wp-post-image\" alt=\"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES\" style=\"aspect-ratio:4\/3;width:100%;object-fit:cover;\" decoding=\"async\" loading=\"lazy\" srcset=\"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png 960w, https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1-300x169.png 300w, https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1-768x432.png 768w\" sizes=\"auto, (max-width: 960px) 100vw, 960px\" \/><\/a><\/figure>\n\n\t\t\n\t\t<div class=\"wp-block-group is-vertical is-layout-flex wp-container-core-group-is-layout-43282307 wp-block-group-is-layout-flex\">\n\t<p style=\"font-style:normal;font-weight:700\" class=\"has-link-color wp-elements-1 wp-block-lp-post-type has-text-color has-primary-color has-small-font-size\">Success story<\/p>\n\n\t\t\n\t\t<h3 style=\"font-style:normal;font-weight:400\" class=\"wp-block-post-title has-base-font-size\"><a href=\"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/\" target=\"_self\" >BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES<\/a><\/h3><\/div>\n\t\t\n\t<\/div>\n\n","yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v28.4 (Yoast SEO v28.4) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES | Devoteam<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES\" \/>\n<meta property=\"og:description\" content=\"Le Bon Coin, Expedia, Glassdoor, Deezer, The New York Times, L\u2019Or\u00e9al\u2026 toutes ces entreprises ont comme point commun de g\u00e9rer un grand volume de pages web. Afin d\u2019optimiser leur r\u00e9f\u00e9rencement dans les moteurs de recherche, elles font appel \u00e0&nbsp;Botify&nbsp;: l\u2019entreprise crawle 5 milliards de pages par mois, soit plus de 2 petabytes de donn\u00e9es. Pour [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/\" \/>\n<meta property=\"og:site_name\" content=\"Devoteam\" \/>\n<meta property=\"article:modified_time\" content=\"2025-02-04T15:08:09+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data1\" content=\"6 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/\",\"url\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/\",\"name\":\"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES | Devoteam\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/devoteam.info\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/Untitled-presentation-1-1.png\",\"datePublished\":\"2023-05-24T20:34:40+00:00\",\"dateModified\":\"2025-02-04T15:08:09+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/#primaryimage\",\"url\":\"https:\\\/\\\/devoteam.info\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/Untitled-presentation-1-1.png\",\"contentUrl\":\"https:\\\/\\\/devoteam.info\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/Untitled-presentation-1-1.png\",\"width\":960,\"height\":540},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Success stories\",\"item\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/success-story\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/#website\",\"url\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/\",\"name\":\"Devoteam\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/devoteam.info\\\/fr\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES | Devoteam","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:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/","og_locale":"fr_FR","og_type":"article","og_title":"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES","og_description":"Le Bon Coin, Expedia, Glassdoor, Deezer, The New York Times, L\u2019Or\u00e9al\u2026 toutes ces entreprises ont comme point commun de g\u00e9rer un grand volume de pages web. Afin d\u2019optimiser leur r\u00e9f\u00e9rencement dans les moteurs de recherche, elles font appel \u00e0&nbsp;Botify&nbsp;: l\u2019entreprise crawle 5 milliards de pages par mois, soit plus de 2 petabytes de donn\u00e9es. Pour [&hellip;]","og_url":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/","og_site_name":"Devoteam","article_modified_time":"2025-02-04T15:08:09+00:00","twitter_card":"summary_large_image","twitter_misc":{"Dur\u00e9e de lecture estim\u00e9e":"6 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/","url":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/","name":"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES | Devoteam","isPartOf":{"@id":"https:\/\/devoteam.info\/fr\/#website"},"primaryImageOfPage":{"@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/#primaryimage"},"image":{"@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/#primaryimage"},"thumbnailUrl":"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png","datePublished":"2023-05-24T20:34:40+00:00","dateModified":"2025-02-04T15:08:09+00:00","breadcrumb":{"@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/#primaryimage","url":"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png","contentUrl":"https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png","width":960,"height":540},{"@type":"BreadcrumbList","@id":"https:\/\/devoteam.info\/fr\/success-story\/botify-passe-a-la-vitesse-superieure-avec-amazon-sagemaker-pipelines\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/devoteam.info\/fr\/"},{"@type":"ListItem","position":2,"name":"Success stories","item":"https:\/\/devoteam.info\/fr\/success-story\/"},{"@type":"ListItem","position":3,"name":"BOTIFY PASSE \u00c0 LA VITESSE SUP\u00c9RIEURE AVEC AMAZON SAGEMAKER PIPELINES"}]},{"@type":"WebSite","@id":"https:\/\/devoteam.info\/fr\/#website","url":"https:\/\/devoteam.info\/fr\/","name":"Devoteam","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/devoteam.info\/fr\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"}]}},"uagb_featured_image_src":{"full":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png",960,540,false],"thumbnail":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1-150x150.png",150,150,true],"medium":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1-300x169.png",300,169,true],"medium_large":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1-768x432.png",768,432,true],"large":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png",960,540,false],"1536x1536":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png",960,540,false],"2048x2048":["https:\/\/devoteam.info\/wp-content\/uploads\/2025\/01\/Untitled-presentation-1-1.png",960,540,false]},"uagb_author_info":{"display_name":"Julien Pawlowski","author_link":"https:\/\/devoteam.info\/fr\/author\/"},"uagb_comment_info":0,"uagb_excerpt":"Le Bon Coin, Expedia, Glassdoor, Deezer, The New York Times, L\u2019Or\u00e9al\u2026 toutes ces entreprises ont comme point commun de g\u00e9rer un grand volume de pages web. Afin d\u2019optimiser leur r\u00e9f\u00e9rencement dans les moteurs de recherche, elles font appel \u00e0&nbsp;Botify&nbsp;: l\u2019entreprise crawle 5 milliards de pages par mois, soit plus de 2 petabytes de donn\u00e9es. Pour\u2026","_links":{"self":[{"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/success-story\/508979","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/success-story"}],"about":[{"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/types\/success-story"}],"version-history":[{"count":0,"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/success-story\/508979\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/media\/509086"}],"wp:attachment":[{"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/media?parent=508979"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/categories?post=508979"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/tags?post=508979"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/devoteam.info\/fr\/wp-json\/wp\/v2\/industry?post=508979"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}