Superpod
PulseAugur coverage of Superpod — every cluster mentioning Superpod across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Huawei's Ascend 950 SuperPoD wins award, surpasses 1,000 deployments
Huawei's Ascend 950 SuperPoD has received the WIC Leading Technology Award. The company announced that its SuperPoD deployments have exceeded 1,000 units, achieving a 45% training utilization rate with its 1,024-NPU system.
-
Wuhan activates 1,024-NPU Ascend 950DT SuperPoD cluster in one month
A new SuperPoD cluster featuring 1,024 Ascend 950DT NPUs has been activated in Wuhan. This liquid-cooled system, capable of 1 EFLOPS of FP8 compute and equipped with 256TB of unified memory, was brought online in just o…
-
OpenEuler aims to unify AI racks into a single compute unit
OpenEuler is developing a system to consolidate an entire rack into a single computing unit. This architecture, featuring SuperPoD and UnifiedBus, aims to enable shared heterogeneous memory and accelerators, alongside a…
-
Huawei unveils Ascend 960 SuperPoD with NPO tech for AI
Huawei has announced its new Ascend 960 SuperPoD, featuring near-package optics (NPO) technology. This advanced infrastructure is designed to support next-generation AI, with a target of approximately 4,096 cards per no…
-
Huawei forecasts AI agents to dominate 90% of global traffic by 2035 · 4 sources tracked
Huawei has released a forecast predicting that AI agents will constitute over 90% of global AI token traffic by 2035. This surge in agent activity is expected to drive a 100,000-fold increase in global computing demand …
-
Huawei unveils Atlas 950 SuperPoD servers with 1,024 NPUs and 256TB memory · 4 sources tracked
Huawei has showcased its Atlas 950 SuperPoD servers, a system comprising 1,024 NPUs designed for AI workloads. The system features extensive inter-rack connectivity, with each of the 32 LRS units utilizing 4x100G connec…
-
NVIDIA unveils first AI-focused Superpod for rapid computation
IEEE Spectrum's Computing Editor, Dina Genkina, was given a tour of NVIDIA's inaugural Superpod. This specialized computing system is built to handle AI workloads, featuring numerous racks of GPUs designed for rapid com…