雷峰网发布于 08/24 16:41

为什么手机内存进入英伟达机柜后,贵过HBM?

过去长期由HBM主导的AI服务器内存账单,正在被“手机内存”改写。 8月11日,外媒Wccftech援引美国银行(BofA)测算称,英伟达面向Vera Rubin Ultra NVL144的Kyber机柜将配置124.4TB HBM4E,成本约为250万美元;同柜216TB LPDDR5X虽然单价更低,但总成本趋近280万美元,超出前者约12%。

查看原文
Microsoft AI Blog发布于 07/21 02:20

Microsoft expands Azure AI and HPC infrastructure with AMD

(翻译)微软借助AMD扩展Azure AI和HPC基础设施

AI workloads are scaling faster than any single infrastructure approach can support — with more models, new agent-driven workloads and surging compute demand driving the need for greater specialization across the stack. To meet this need, Microsoft continues to evolve Azure’s infrastructure, includi

查看原文
NVIDIA Technical Blog发布于 08/24 23:00

Maximizing AI Factory Performance per Watt with NVIDIA DSX MaxLPS

(翻译)借助 NVIDIA DSX MaxLPS 最大化 AI 工厂每瓦性能

AI factories are power-constrained industrial systems. The question is no longer how many GPUs fit in a data center, but how much AI output each available…

查看原文
NVIDIA Technical Blog发布于 08/24 23:00

Solving Agentic AI Fleet Challenges with NVIDIA Vera CPU

(翻译)利用 NVIDIA Vera CPU 解决 Agentic AI 集群挑战

AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks.

查看原文