‘AI success will be defined not by how much infrastructure organizations own, but by how productively they use it’: Nvidia lays out its thoughts on how storage has become the next frontier of AI

Nvidia 将 cuFile 存储接口及底层存储栈开源,并拉上 Google、Intel、Meta 等 40 多家厂商成立 Storage-Next 联盟,核心目标是让 GPU 绕过 CPU 直接管理存储 I/O,以解决 AI 推理中 KV Cache 带来的海量小数据读取瓶颈。








