NVIDIA DSX MaxLPS 提升 AI 工厂吞吐与能效
原标题：How NVIDIA DSX MaxLPS Maximizes AI Factory Throughput and Efficiency
AI 摘要
NVIDIA 发布技术博客，介绍 DSX MaxLPS 动态功耗共享方案，通过策略化分配在固定电力预算内提升 AI 工厂吞吐。NVIDIA 与 Nscale 在冰岛 Keflavík 数据中心使用 GB300 NVL72 系统、Kimi K2.5 FP4 等进行了联合评估，结果显示在相同 264.4 kW 电力预算下，GPU 数量从 140 增至 192（+37.1%），聚合吞吐提升 49.2%，每瓦吞吐从 4.10 提升至 6.12 tokens/s/W。
正文节选
Every unused watt is capacity left on the table. AI factories are typically provisioned for the unlikely moment when every GPU reaches peak power, creating a protective buffer that can leave valuable infrastructure underused during normal operation. NVIDIA DSX MaxLPS uses policy-governed power sharing to allocate power across participating resources dynamically, enabling customers to deploy up to 40% more GPUs within the same approved power budget. This technical walkthrough examines a joint NVI