Ascend Atlas 950 SuperPoD

Headlines · 1

1. Oct 23
   NVIDIA DSX POD targets faster Modular AI Data-Center Deployment

   36Kr reported that NVIDIA announced DSX POD, a modular reference architecture for the Vera Rubin platform that integrates computing equipment, rack power, networking and liquid cooling for on-site delivery. NVIDIA says deployment can fall from as long as six months to as little as 14 days when external power, cooling and networking are already prepared; partner products are expected from early 2027. The report also reviews modular infrastructure efforts from ZTE and Huawei, a reported 1,024-accelerator Ascend Atlas 950 SuperPoD milestone, and NAND production-line upgrades by Samsung and SK Hynix in China.

   36氪 人工智能 · 🔥 0

Experience and discussion from the community

Share my Ascend Atlas 950 SuperPoD experienceAsk about Ascend Atlas 950 SuperPoD