{"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\/sq\/blog\/news\/proekt-zluda-prodolzhit-razvitie-v-forme-universalnoj-realizaczii-cuda","title":{"rendered":"Projekti ZLUDA do t\u00eb vazhdoj\u00eb zhvillimin si nj\u00eb realizim universitar i CUDA","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
Andrzej Janik ka paraqitur planin p\u00ebr zhvillimin e m\u00ebtejsh\u00ebm t\u00eb projektit ZLUDA, q\u00eb zhvillon nj\u00eb implementim t\u00eb hapur t\u00eb teknologjis\u00eb CUDA. N\u00ebse varianti i par\u00eb i ZLUDA kishte si q\u00ebllim krijimin e nj\u00eb implementimi t\u00eb CUDA p\u00ebr GPU-t\u00eb Intel, dhe varianti i dyt\u00eb p\u00ebr GPU-t\u00eb AMD, varianti i tret\u00eb i paraqitur tani do t\u00eb p\u00ebrqendrohet n\u00eb zhvillimin e nj\u00eb implementimi universale t\u00eb CUDA p\u00ebr \u00e7do GPU q\u00eb nuk \u00ebsht\u00eb NVIDIA. Q\u00ebllimi i projektit ZLUDA \u00ebsht\u00eb t\u00eb ofroj\u00eb n\u00eb sistemet me GPU, t\u00eb ndryshme nga ato NVIDIA, mund\u00ebsin\u00eb p\u00ebr t\u00eb ekzekutuar aplikacione CUDA t\u00eb pa modifikuara me nj\u00eb performanc\u00eb t\u00eb ngjashme me at\u00eb t\u00eb aplikacioneve q\u00eb ekzekutohen pa nd\u00ebrmjet\u00ebs. Kodi i projektit vazhdon t\u00eb shp\u00ebrndahet n\u00ebn licencat MIT dhe Apache 2.0. <\/p>\n
Fillimisht, projekti ZLUDA u zhvillua p\u00ebr GPU Intel, por n\u00eb vitin 2021, kjo kompani e konsideroi ofrimin e mund\u00ebsis\u00eb p\u00ebr t\u00eb drejtuar aplikacione CUDA n\u00eb GPU Intel si t\u00eb pap\u00ebrshtatshme p\u00ebr biznesin dhe nuk vazhdoi mb\u00ebshtetje p\u00ebr zhvillimin e iniciativ\u00ebs. N\u00eb fillim t\u00eb vitit 2022, zhvilluesi u largua nga Intel dhe n\u00ebnshkroi nj\u00eb kontrat\u00eb me AMD p\u00ebr t\u00eb zhvilluar nj\u00eb shtres\u00eb p\u00ebr kompatibilitetin me CUDA p\u00ebr k\u00ebt\u00eb kompani. Pas dy vjet\u00ebsh, kompania AMD humbi interesin p\u00ebr mund\u00ebsin\u00eb e ekzekutimit t\u00eb aplikacioneve CUDA n\u00eb GPU AMD. <\/p>\n
N\u00eb p\u00ebrputhje me kushtet e kontrat\u00ebs dhe pas marrjes s\u00eb lejes p\u00ebr publikim nga nj\u00eb p\u00ebrfaq\u00ebsues i AMD gjat\u00eb nj\u00eb komunikimi p\u00ebrmes email-it, Andrzej publikoi kodin e punimeve, t\u00eb krijuar gjat\u00eb pun\u00ebs n\u00eb AMD dhe q\u00eb mund\u00ebsonin ekzekutimin e aplikacioneve CUDA mbi stakun ROCm dhe runtime HIP (Heterogeneous-computing Interface for Portability) t\u00eb zhvilluar nga AMD. Gjasht\u00eb muaj pas publikimit, avokat\u00ebt e AMD kontaktuan Andrzejn dhe shprehnin se leja e dh\u00ebn\u00eb m\u00eb par\u00eb gjat\u00eb komunikimit nuk ka fuqi ligjore. Si rezultat, Andrzej u detyrua t\u00eb t\u00ebrheq\u00eb nga qasja publike kodin e ZLUDA-s, t\u00eb krijuar gjat\u00eb koh\u00ebs s\u00eb pun\u00ebs n\u00eb AMD. <\/p>\n
Varianti i ri i ZLUDA do t\u00eb bazohet n\u00eb kodin burimor q\u00eb ekzistonte p\u00ebrpara fillimit t\u00eb pun\u00ebs s\u00eb Andrzejt n\u00eb AMD, nuk do t\u00eb jet\u00eb i lidhur me GPU t\u00eb caktuar dhe do t\u00eb p\u00ebrqendrohet n\u00eb ekzekutimin e aplikacioneve q\u00eb p\u00ebrdorin CUDA p\u00ebr t\u00eb p\u00ebrshpejtuar detyrat e lidhura me m\u00ebsimin e makinerive dhe inteligjenc\u00ebn artificiale (m\u00eb par\u00eb, ZLUDA ishte kryesisht i fokusuar n\u00eb aplikacionet p\u00ebr krijimin e p\u00ebrmbajtjes, p\u00ebrfshir\u00eb Arnold Render, Blender dhe 3DF Zephyr). Nd\u00ebr t\u00eb tjera, planifikohet q\u00eb t\u00eb ofrohet mund\u00ebsia e pun\u00ebs s\u00eb kuadrove Llama.cpp, PyTorch dhe TensorFlow mbi ZLUDA duke angazhuar optimizimet CUDA t\u00eb aplikuara p\u00ebr GPU-t\u00eb NVIDIA. <\/p>\n
N\u00eb fillim, projekti do t\u00eb fokusohet n\u00eb pun\u00ebn me GPU-t\u00eb AMD, por m\u00eb von\u00eb do t\u00eb adaptohet p\u00ebr GPU-t\u00eb Intel. Zbatimi p\u00ebr GPU-t\u00eb AMD do t\u00eb nd\u00ebrtohet nga e para, do t\u00eb mb\u00ebshtes\u00eb GPU-t\u00eb, t\u00eb pakt\u00ebn t\u00eb bazuara n\u00eb arkitektur\u00ebn e grupeve t\u00eb komandave RDNA1, dhe do t\u00eb p\u00ebrdor\u00eb paket\u00ebn ROCm 6.1+ n\u00eb vend t\u00eb paket\u00ebs ROCm 5, e cila u p\u00ebrdor n\u00eb zbatimin e m\u00ebparsh\u00ebm p\u00ebr AMD. Pariteti i funksionalitetit me zbatimin e kaluar, t\u00eb cilin Andrzej e detyroi t\u00eb hiqte nga akses i hapur, planifikohet t\u00eb arrihet n\u00eb tremujorin e tret\u00eb t\u00eb vitit 2025. \n Burimi: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"