<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>💾 ZeRO on Elon&#39;s AD Insight</title>
    <link>https://auto-driving-blog.vercel.app/tags/-zero/</link>
    <description>Recent content in 💾 ZeRO on Elon&#39;s AD Insight</description>
    <image>
      <title>Elon&#39;s AD Insight</title>
      <url>https://auto-driving-blog.vercel.app/images/share.png</url>
      <link>https://auto-driving-blog.vercel.app/images/share.png</link>
    </image>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Sun, 19 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://auto-driving-blog.vercel.app/tags/-zero/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>分布式训练的工程实践：从DeepSpeed到3D并行</title>
      <link>https://auto-driving-blog.vercel.app/posts/knowledge/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E7%9A%84%E5%B7%A5%E7%A8%8B%E5%AE%9E%E8%B7%B5/</link>
      <pubDate>Sun, 19 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://auto-driving-blog.vercel.app/posts/knowledge/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E7%9A%84%E5%B7%A5%E7%A8%8B%E5%AE%9E%E8%B7%B5/</guid>
      <description>大模型训练需要分布式系统支持，但3D并行（数据并行/张量并行/流水线并行）的配置与调优极为复杂。本文从单卡训练痛点出发，系统讲解DeepSpeed的ZeRO stages、Megatron-LM的张量切片策略、流水线并行的micro-batch调度，以及VLA模型训练中activation checkpointing与通信拓扑选择的工程经验。</description>
    </item>
  </channel>
</rss>
