NVIDIA Gives First Look at New Supercomputer

<p>NVIDIA has released a video that offers the first public look at Eos&comma; its latest data-center-scale supercomputer&period;<&sol;p>&NewLine;<p>An extremely large-scale <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;data-center&sol;dgx-superpod&sol;">NVIDIA DGX SuperPOD<&sol;a>&comma; Eos is where NVIDIA developers create their AI breakthroughs using accelerated computing infrastructure and fully optimized software&period;<&sol;p>&NewLine;<p>Eos is built with 576 <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;data-center&sol;dgx-h100&sol;">NVIDIA DGX H100<&sol;a> systems&comma; <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;networking&sol;quantum2&sol;">NVIDIA Quantum-2 InfiniBand networking<&sol;a> and software&comma; providing a total of 18&period;4 exaflops of FP8 AI performance&period;<&sol;p>&NewLine;<p>Revealed in November at the Supercomputing 2023 trade show&comma; Eos — named for the Greek goddess said to open the gates of dawn each day — reflects NVIDIA’s commitment to advancing AI technology&period;<&sol;p>&NewLine;<p>Each DGX H100 system is equipped with eight <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;data-center&sol;h100&sol;">NVIDIA H100 Tensor Core GPUs<&sol;a>&period; Eos features a total of 4&comma;608 H100 GPUs&period;<&sol;p>&NewLine;<p>As a result&comma; Eos can handle the largest AI workloads to train <a href&equals;"https&colon;&sol;&sol;blogs&period;nvidia&period;com&sol;blog&sol;kt-large-language-models&sol;">large language models<&sol;a>&comma; recommender systems&comma; quantum simulations and more&period;<&sol;p>&NewLine;<p>It’s a showcase of what NVIDIA’s technologies can do&comma; when working at scale&period;<&sol;p>&NewLine;<p>Eos is arriving at the perfect time&period; People are changing the world with <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;glossary&sol;generative-ai&sol;">generative AI<&sol;a>&comma; from drug discovery to chatbots to autonomous machines and beyond&period;<&sol;p>&NewLine;<p>To achieve these breakthroughs&comma; they need more than AI expertise and development skills&period; They need an AI factory — a purpose-built AI engine that’s always available and can help ramp their capacity to build AI models at scale<&sol;p>&NewLine;<p>Eos is r<a href&equals;"https&colon;&sol;&sol;www&period;top500&period;org&sol;lists&sol;top500&sol;2023&sol;11&sol;">anked No&period; 9 in the TOP500<&sol;a> list of the world’s fastest supercomputers&period;<&sol;p>&NewLine;<p>It includes NVIDIA’s advanced accelerated computing and networking alongside sophisticated software offerings such as <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;data-center&sol;base-command&sol;">NVIDIA Base Command<&sol;a> and <a href&equals;"https&colon;&sol;&sol;www&period;nvidia&period;com&sol;en-us&sol;data-center&sol;products&sol;ai-enterprise&sol;">NVIDIA AI Enterprise<&sol;a>&period;<&sol;p>&NewLine;<p>Eos’s architecture is optimized for AI workloads demanding ultra-low-latency and high-throughput interconnectivity across a large cluster of accelerated computing nodes&comma; making it an ideal solution for enterprises looking to scale their AI capabilities&period;<&sol;p>&NewLine;<p>Based on NVIDIA Quantum-2 InfiniBand with In-Network Computing technology&comma; its network architecture supports data transfer speeds of up to 400Gb&sol;s&comma; facilitating the rapid movement of large datasets essential for training complex AI models&period;<&sol;p>&NewLine;<p>At the heart of Eos lies the groundbreaking DGX SuperPOD architecture powered by NVIDIA’s DGX H100 systems&period;<&sol;p>&NewLine;<p>The architecture is built to provide the AI and computing fields with tightly integrated full-stack systems capable of computing at an enormous scale&period;<&sol;p>&NewLine;<p><iframe width&equals;"560" height&equals;"315" src&equals;"https&colon;&sol;&sol;www&period;youtube&period;com&sol;embed&sol;J8-CgG5ewJQ&quest;si&equals;Aooz5uwQcHCxxApq" title&equals;"YouTube video player" frameborder&equals;"0" allow&equals;"accelerometer&semi; autoplay&semi; clipboard-write&semi; encrypted-media&semi; gyroscope&semi; picture-in-picture&semi; web-share" allowfullscreen><&sol;iframe><&sol;p>&NewLine;

Editor

Editor

Wispr Scores $25 Million Series A Extension

SAN FRANCISCO -- Wispr, the voice-to-text AI that turns speech into clear, polished writing in every…

1 day

Numeric Dials Up $51 Million Series B

SAN FRANCISCO -- Numeric, an AI accounting automation platform, has raised a $51 million Series…

1 day

Apple Names 45 Finalists for App Store of the Year Awards

Apple has announced 45 finalists for this year’s App Store Awards, recognizing the best apps…

2 days

UC Reaches Agreement With Nurses, Strike Canceled

The University of California (UC) and the California Nurses Association (CNA) have reached a tentative…

4 days

HouseRX Rakes In $55 Million Series B

SAN FRANCISCO -- House Rx, a health tech company focused on making specialty medications more accessible and…

4 days

King Charles Honors NVIDIA’s Jensen Huang

Britain's King has given an award to the King of NVIDIA! NVIDIA founder and CEO…

4 days