# Nvidia published DWDP （Distributed Weight-Data Parallelism）， a new inference parallelism strategy fo…

- 来源：SemiAnalysis (@SemiAnalysis_)
- 发布时间：2026-04-10 01:00
- AIHOT 链接：https://aihot.virxact.com/items/cmnxjn75l00dfsl9og3ekvybt
- 原文链接：https://x.com/SemiAnalysis_/status/2042286547769184644

## AI 摘要

Nvidia 发布了 DWDP (Distributed Weight-Data Parallelism)，这是一种专注于 prefill 的新推理并行策略。这听起来有点疯狂，直到你想起目标机器是 GB200 NVL72。核心权衡：花费更多 peer-GPU 带宽，从而减少在 collective barriers 上的等待时间。(1/6) 🧵 https://arxiv.org/abs/2604.01621v1

## 正文

Nvidia published DWDP （Distributed Weight-Data Parallelism）， a new inference parallelism strategy focused on prefill. It sounds slightly insane until you remember the target machine is GB200 NVL72. The core trade： spend more peer-GPU bandwidth so you spend less time waiting at collective barriers. （1/6） 🧵 https://arxiv.org/abs/2604.01621v1