{"id":120066,"date":"2024-10-07T01:09:14","date_gmt":"2024-10-06T23:09:14","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda"},"modified":"2024-10-07T01:09:14","modified_gmt":"2024-10-06T23:09:14","slug":"proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/news\/proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda","title":{"rendered":"Projekt ZLUDA bude pokra\u010dovat ve v\u00fdvoji jako univerz\u00e1ln\u00ed implementace CUDA","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Andrzej Janik p\u0159edstavil pl\u00e1n dal\u0161\u00edho rozvoje projektu ZLUDA, kter\u00fd vyv\u00edj\u00ed otev\u0159enou implementaci technologie CUDA. Pokud prvn\u00ed varianta ZLUDA byla zam\u011b\u0159ena na vytvo\u0159en\u00ed implementace CUDA pro GPU Intel a druh\u00e1 - pro GPU AMD, nyn\u00ed p\u0159edstaven\u00e1 t\u0159et\u00ed varianta se zam\u011b\u0159\u00ed na rozvoj univerz\u00e1ln\u00ed implementace CUDA pro jak\u00e9koli GPU, krom\u011b GPU NVIDIA. C\u00edlem projektu ZLUDA je poskytnout na syst\u00e9mech s GPU, kter\u00e9 nejsou od NVIDIA, mo\u017enost spou\u0161t\u011bn\u00ed nemodifikovan\u00fdch aplikac\u00ed CUDA s v\u00fdkonem bl\u00edzk\u00fdm v\u00fdkonu aplikac\u00ed b\u011b\u017e\u00edc\u00edch bez vrstev. K\u00f3d projektu se i nad\u00e1le \u0161\u00ed\u0159\u00ed pod licencemi MIT a Apache 2.0. <\/p>\n

P\u016fvodn\u011b byl projekt ZLUDA vyv\u00edjen pro GPU Intel, ale v roce 2021 tato spole\u010dnost dosp\u011bla k z\u00e1v\u011bru, \u017ee mo\u017enost spou\u0161t\u011bt CUDA aplikace na GPU Intel nen\u00ed pro podnik\u00e1n\u00ed zaj\u00edmav\u00e1, a nepodporovala dal\u0161\u00ed rozvoj iniciativy. Na po\u010d\u00e1tku roku 2022 se v\u00fdvoj\u00e1\u0159 vzd\u00e1lil od Intelu a podepsal smlouvu s AMD na v\u00fdvoj vrstvy pro kompatibilitu s CUDA pro GPU t\u00e9to spole\u010dnosti. Po dvou letech ztratila spole\u010dnost AMD z\u00e1jem o mo\u017enost spou\u0161t\u011bn\u00ed CUDA aplikac\u00ed na GPU AMD. <\/p>\n

V souladu s podm\u00ednkami smlouvy a po obdr\u017een\u00ed povolen\u00ed k publikov\u00e1n\u00ed od z\u00e1stupce AMD b\u011bhem e-mailov\u00e9 korespondence Andrzej zve\u0159ejnil k\u00f3d prac\u00ed vyvinut\u00fdch b\u011bhem sv\u00e9ho p\u016fsoben\u00ed v AMD, kter\u00e9 umo\u017e\u0148uj\u00ed spou\u0161t\u011bn\u00ed CUDA aplikac\u00ed nad vyv\u00edjenou technologi\u00ed ROCm a runtime HIP (Heterogeneous-computing Interface for Portability) od AMD. \u0160est m\u011bs\u00edc\u016f po publikaci kontaktovali pr\u00e1vn\u00edci AMD Andrzeje a dali najevo, \u017ee d\u0159\u00edve poskytnut\u00e9 povolen\u00ed v pr\u016fb\u011bhu korespondence nem\u00e1 pr\u00e1vn\u00ed s\u00edlu. V d\u016fsledku toho byl Andrzej nucen st\u00e1hnout z ve\u0159ejn\u00e9ho p\u0159\u00edstupu k\u00f3d ZLUDA, kter\u00fd byl vytvo\u0159en b\u011bhem jeho pr\u00e1ce v AMD. <\/p>\n

Nov\u00e1 verze ZLUDA bude vych\u00e1zet z k\u00f3dov\u00e9 z\u00e1kladny, kter\u00e1 existovala p\u0159ed za\u010d\u00e1tkem pr\u00e1ce Andrzeje na AMD, nebude v\u00e1z\u00e1na na \u017e\u00e1dn\u00e9 konkr\u00e9tn\u00ed GPU a zam\u011b\u0159\u00ed se na prov\u00e1d\u011bn\u00ed aplikac\u00ed vyu\u017e\u00edvaj\u00edc\u00edch CUDA pro zrychlen\u00ed \u00faloh souvisej\u00edc\u00edch s strojov\u00fdm u\u010den\u00edm a um\u011blou inteligenc\u00ed (d\u0159\u00edve byl ZLUDA zam\u011b\u0159en p\u0159edev\u0161\u00edm na aplikace pro tvorbu obsahu, jako jsou Arnold Render, Blender a 3DF Zephyr). Mezi jin\u00fdmi se pl\u00e1nuje poskytnout mo\u017enost fungov\u00e1n\u00ed framework\u016f Llama.cpp, PyTorch a TensorFlow nad ZLUDA s vyu\u017eit\u00edm CUDA optimalizac\u00ed aplikovan\u00fdch na GPU NVIDIA. <\/p>\n

Zpo\u010d\u00e1tku se projekt zam\u011b\u0159\u00ed na pr\u00e1ci na GPU AMD, ale pot\u00e9 bude p\u0159izp\u016fsoben pro GPU Intel. Implementace pro GPU AMD bude vytvo\u0159ena od z\u00e1klad\u016f, podpo\u0159\u00ed GPU minim\u00e1ln\u011b na z\u00e1klad\u011b architektury sady instrukc\u00ed RDNA1, a bude pou\u017e\u00edvat stack ROCm 6.1+ m\u00edsto stacku ROCm 5, kter\u00fd byl pou\u017e\u00edv\u00e1n v d\u0159\u00edve vytvo\u0159en\u00e9 implementaci pro AMD. Parity funkc\u00ed s p\u0159edchoz\u00ed implementac\u00ed, kterou byl Andrzej nucen st\u00e1hnout z ve\u0159ejn\u00e9ho p\u0159\u00edstupu, by m\u011blo b\u00fdt dosa\u017eeno ve t\u0159et\u00edm \u010dtvrtlet\u00ed 2025.
\n
Zdroj: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u0410\u043d\u0434\u0436\u0435\u0439 \u042f\u043d\u0438\u043a (Andrzej Janik) \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u0438\u043b \u043f\u043b\u0430\u043d \u0434\u0430\u043b\u044c\u043d\u0435\u0439\u0448\u0435\u0439 \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0438 \u043f\u0440\u043e\u0435\u043a\u0442\u0430 ZLUDA, \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u044e\u0449\u0435\u0433\u043e \u043e\u0442\u043a\u0440\u044b\u0442\u0443\u044e \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u044e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 CUDA. \u0415\u0441\u043b\u0438 \u043f\u0435\u0440\u0432\u044b\u0439 \u0432\u0430\u0440\u0438\u0430\u043d\u0442 ZLUDA \u0431\u044b\u043b \u043d\u0430\u0446\u0435\u043b\u0435\u043d \u043d\u0430 \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0435 \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438 CUDA \u0434\u043b\u044f GPU Intel, \u0430 \u0432\u0442\u043e\u0440\u043e\u0439 — \u0434\u043b\u044f GPU AMD, \u0442\u043e \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d\u043d\u044b\u0439 \u0441\u0435\u0439\u0447\u0430\u0441 \u0442\u0440\u0435\u0442\u0438\u0439 \u0432\u0430\u0440\u0438\u0430\u043d\u0442 \u0431\u0443\u0434\u0435\u0442 \u0441\u043e\u0441\u0440\u0435\u0434\u043e\u0442\u043e\u0447\u0435\u043d \u043d\u0430 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u0438 \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0430\u043b\u044c\u043d\u043e\u0439 \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438 CUDA \u0434\u043b\u044f \u043b\u044e\u0431\u044b\u0445 GPU, \u043e\u0442\u043b\u0438\u0447\u043d\u044b\u0445 \u043e\u0442 GPU NVIDIA. \u0426\u0435\u043b\u044c\u044e \u043f\u0440\u043e\u0435\u043a\u0442\u0430 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-120066","post","type-post","status-publish","format-standard","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Projekt ZLUDA bude pokra\u010dovat v rozvoji ve form\u011b univerz\u00e1ln\u00ed realizace CUDA | ProHoster","description":"Andrzej Janik p\u0159edstavil pl\u00e1n dal\u0161\u00edho v\u00fdvoje projektu ZLUDA, kter\u00fd rozv\u00edj\u00ed otev\u0159enou realizaci technologie CUDA.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/news\/proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"cs_CZ","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u041f\u0440\u043e\u0435\u043a\u0442 ZLUDA \u043f\u0440\u043e\u0434\u043e\u043b\u0436\u0438\u0442 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u0435 \u0432 \u0444\u043e\u0440\u043c\u0435 \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0430\u043b\u044c\u043d\u043e\u0439 \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438 CUDA | ProHoster","og:description":"\u0410\u043d\u0434\u0436\u0435\u0439 \u042f\u043d\u0438\u043a (Andrzej Janik) \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u0438\u043b \u043f\u043b\u0430\u043d \u0434\u0430\u043b\u044c\u043d\u0435\u0439\u0448\u0435\u0439 \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0438 \u043f\u0440\u043e\u0435\u043a\u0442\u0430 ZLUDA, \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u044e\u0449\u0435\u0433\u043e \u043e\u0442\u043a\u0440\u044b\u0442\u0443\u044e \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u044e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 CUDA.","og:url":"https:\/\/prohoster.info\/cs\/blog\/news\/proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2024-10-06T23:09:14+00:00","article:modified_time":"2024-10-06T23:09:14+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"120066","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":"default","schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-23 07:06:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2026-01-23 07:06:19","updated":"2026-01-23 07:06:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/120066","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/comments?post=120066"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/120066\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=120066"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=120066"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=120066"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}