Fosstodon

0 posts0 participants0 posts today

Andrew Jones (hpcnotes)From the other place:<a href="https://mast.hpc.social/tags/OpenAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#OpenAI</a> starts using first few <a href="https://mast.hpc.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> racks in <a href="https://mast.hpc.social/tags/Azure" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Azure</a> <a href="https://mast.hpc.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://mast.hpc.social/tags/GPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GPU</a>

Benjamin Carr, Ph.D. 👨🏻‍💻🧬<a href="https://hachyderm.io/tags/HPE" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#HPE</a> Upgrades <a href="https://hachyderm.io/tags/Supercomputer" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Supercomputer</a> Lineup Top To Bottom In #2025 <a href="https://hachyderm.io/tags/Slingshot400" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Slingshot400</a> with s a new “Rosetta-2” switch ASIC and a new “Cassini-2” network adapter ASIC. <a href="https://hachyderm.io/tags/CrayEX" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#CrayEX</a> and <a href="https://hachyderm.io/tags/ProLiantXD" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ProLiantXD</a> upgrades Zen5 and Zen5c with up to 98,304 cores in a single liquid-cooled rack <a href="https://hachyderm.io/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> NVL4 board will allow up to 224 <a href="https://hachyderm.io/tags/Blackwell" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Blackwell</a> <a href="https://hachyderm.io/tags/GPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GPU</a> in a single cabinet 7 of the top ten supercomputers ranked on the November 2024 <a href="https://hachyderm.io/tags/Top500" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Top500</a> rankings are based on the <a href="https://hachyderm.io/tags/Cray" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Cray</a> EX platform; 10 of top 20 machines are too! <a href="https://www.nextplatform.com/2024/11/26/hpe-upgrades-supercomputer-lineup-top-to-bottom-in-2025/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.nextplatform.com/2024/11/26/hpe-upgrades-supercomputer-lineup-top-to-bottom-in-2025/</a>

WinbuzzerThermal issues with Nvidia's Blackwell GPUs force multiple design revisions and disrupt deployment timelines for AI projects. <a href="https://mastodon.social/tags/nvidia" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#nvidia</a> <a href="https://mastodon.social/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a> <a href="https://mastodon.social/tags/blackwell" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#blackwell</a> <a href="https://mastodon.social/tags/B200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#B200</a> <a href="https://mastodon.social/tags/B100" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#B100</a> <a href="https://mastodon.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a><a href="https://buff.ly/48ThCpq" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://buff.ly/48ThCpq</a>

IT NewsAMD unveils powerful new AI chip to challenge Nvidia - On Thursday, AMD announced its new MI325X AI accelerator chip, which is se... - <a href="https://arstechnica.com/ai/2024/10/amd-unveils-powerful-new-ai-chip-to-challenge-nvidia/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://arstechnica.com/ai/2024/10/amd-unveils-powerful-new-ai-chip-to-challenge-nvidia/</a> <a href="https://schleuss.online/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://schleuss.online/tags/aiaccelerator" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#aiaccelerator</a> <a href="https://schleuss.online/tags/nvidiah200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#nvidiah200</a> <a href="https://schleuss.online/tags/biz" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#biz</a>⁢ <a href="https://schleuss.online/tags/mi325x" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#mi325x</a> <a href="https://schleuss.online/tags/nvidia" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#nvidia</a> <a href="https://schleuss.online/tags/gb200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#gb200</a> <a href="https://schleuss.online/tags/mi350" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#mi350</a> <a href="https://schleuss.online/tags/gpus" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#gpus</a> <a href="https://schleuss.online/tags/h200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#h200</a> <a href="https://schleuss.online/tags/amd" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#amd</a> <a href="https://schleuss.online/tags/gpu" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#gpu</a> <a href="https://schleuss.online/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a>

HPC GuruFoxconn is building the world's largest Nvidia <a href="https://mastodon.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> production facility to help meet "awfully huge" demand for the <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> darling's <a href="https://mastodon.social/tags/Blackwell" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Blackwell</a> platformHonHai_Foxconn will be responsible for the <a href="https://mastodon.social/tags/GPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GPU</a> module packaging<a href="https://finance.yahoo.com/news/foxconn-says-building-worlds-largest-024924787.html" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://finance.yahoo.com/news/foxconn-says-building-worlds-largest-024924787.html</a><a href="https://mastodon.social/tags/HPC" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#HPC</a>

HabrGPU для дата-центровСерверный узел с восемью GPU-ускорителями AMD Instinct MI325X (2,3 ТБ видеопамяти, 8×750 Вт) Современные дата-центры уже не такие, как прежде. Раньше на типичный сервер поставили бы 144-ядерный CPU , много RAM и десяток HDD/SSD. Но теперь приоритеты меняются, в первую очередь из-за высокого спроса на машинное обучение (ML) и приложения LLM. Хотя аппетит к памяти только вырос, но процессоры нужны другие. И крупным корпорациям, и маленьким компаниям нужно железо для обсчёта ИИ-приложений. Это GPU-ускорители, NPU (Neural Processing Unit) и TPU (Tensor Processing Unit), AI-чипы нового поколения. Нужны серверы с GPU. Открываются даже специализированные GPU-облака на растущем спросе.<a href="https://habr.com/ru/companies/ruvds/articles/837286/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://habr.com/ru/companies/ruvds/articles/837286/</a><a href="https://zhub.link/tags/GPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GPU</a> <a href="https://zhub.link/tags/NPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#NPU</a> <a href="https://zhub.link/tags/TPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#TPU</a> <a href="https://zhub.link/tags/ARM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ARM</a> <a href="https://zhub.link/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://zhub.link/tags/%D0%98%D0%98" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ИИ</a> <a href="https://zhub.link/tags/Copilot" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Copilot</a>+ <a href="https://zhub.link/tags/Qualcomm_Snapdragon" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Qualcomm_Snapdragon</a> X <a href="https://zhub.link/tags/Core_Ultra" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Core_Ultra</a> <a href="https://zhub.link/tags/AMD_XDNA" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AMD_XDNA</a> <a href="https://zhub.link/tags/Intel_AI_Boost" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Intel_AI_Boost</a> <a href="https://zhub.link/tags/Qualcomm_Hexagon" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Qualcomm_Hexagon</a> <a href="https://zhub.link/tags/Instinct_MI325X" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Instinct_MI325X</a> <a href="https://zhub.link/tags/Nvidia_H100" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia_H100</a> <a href="https://zhub.link/tags/Nvidia_H200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia_H200</a> <a href="https://zhub.link/tags/Nvidia_GH100" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia_GH100</a> <a href="https://zhub.link/tags/Instinct_MI300X" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Instinct_MI300X</a> <a href="https://zhub.link/tags/Nvidia_B200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia_B200</a> <a href="https://zhub.link/tags/Grace_Blackwell_Superchip" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Grace_Blackwell_Superchip</a> <a href="https://zhub.link/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> <a href="https://zhub.link/tags/AMD_Ryzen_AI_300" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AMD_Ryzen_AI_300</a> <a href="https://zhub.link/tags/MediaTek" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#MediaTek</a> <a href="https://zhub.link/tags/OpenVINO" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#OpenVINO</a> <a href="https://zhub.link/tags/ruvds_%D1%81%D1%82%D0%B0%D1%82%D1%8C%D0%B8" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ruvds_статьи</a>

Digital Human ✔Jim keller's advies aan nvidia: gebruik ethernet voor gb200 gpu-connectiviteit <a href="https://www.trendingtech.news/trending-news/2024/04/6524/jim-keller-s-advies-aan-nvidia-gebruik-ethernet-voor-gb200-gpu-connectiviteit" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.trendingtech.news/trending-news/2024/04/6524/jim-keller-s-advies-aan-nvidia-gebruik-ethernet-voor-gb200-gpu-connectiviteit</a> <a href="https://mastodon.social/tags/Nvidia" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia</a> <a href="https://mastodon.social/tags/Jim" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Jim</a> Keller <a href="https://mastodon.social/tags/Ethernet" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Ethernet</a> <a href="https://mastodon.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> GPU <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://mastodon.social/tags/Trending" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Trending</a> <a href="https://mastodon.social/tags/News" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#News</a> <a href="https://mastodon.social/tags/Nieuws" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nieuws</a>

eicker.news ᳇ tech news»<a href="https://eicker.news/tags/Nvidia" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Nvidia</a>’s must-have <a href="https://eicker.news/tags/H100" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#H100</a> <a href="https://eicker.news/tags/AIchip" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AIchip</a> made it a multitrillion-dollar company, but perhaps Nvidia is about to extend its <a href="https://eicker.news/tags/lead" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#lead</a> — with the new <a href="https://eicker.news/tags/Blackwell" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Blackwell</a> <a href="https://eicker.news/tags/B200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#B200</a> <a href="https://eicker.news/tags/GPU" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GPU</a> and <a href="https://eicker.news/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> “<a href="https://eicker.news/tags/superchip" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#superchip</a>.”« <a href="https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai?eicker.news" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai?eicker.news</a> <a href="https://eicker.news/tags/tech" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#tech</a> <a href="https://eicker.news/tags/media" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#media</a>

Glenn K. Lockwood<a href="https://mast.hpc.social/tags/Azure" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Azure</a> will be launching <a href="https://mast.hpc.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> with 800G InfiniBand (<a href="https://azure.microsoft.com/en-us/blog/microsoft-and-nvidia-partnership-continues-to-deliver-on-the-promise-of-ai/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://azure.microsoft.com/en-us/blog/microsoft-and-nvidia-partnership-continues-to-deliver-on-the-promise-of-ai/</a>). Superchips and SuperNICs. Everything is super.

FCLCEhhhh the newest big GPU has arrived!And you can have two of them connected to Grace! <a href="https://mast.hpc.social/tags/B100" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#B100</a> and <a href="https://mast.hpc.social/tags/GB200" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GB200</a> for the GPU itself and the <a href="https://mast.hpc.social/tags/grace" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#grace</a> +2X GPU version <a href="https://mast.hpc.social/tags/hpc" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#hpc</a> <a href="https://mast.hpc.social/tags/GTC" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GTC</a>

Recent searches

Search options

Administered by:

Server stats:

#gb200