<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>NAND闪存 on 知识铺的博客</title>
    <link>https://index.zshipu.com/ai001/tags/NAND%E9%97%AA%E5%AD%98/</link>
    <description>Recent content in NAND闪存 on 知识铺的博客</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh-CN</language>
    <lastBuildDate>Wed, 02 Sep 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://index.zshipu.com/ai001/tags/NAND%E9%97%AA%E5%AD%98/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>美光近GPU NAND闪存让数百GB存储直达GPU</title>
      <link>https://index.zshipu.com/ai001/post/20260902/%E7%BE%8E%E5%85%89%E8%BF%91GPU-NAND%E9%97%AA%E5%AD%98%E8%AE%A9%E6%95%B0%E7%99%BEGB%E5%AD%98%E5%82%A8%E7%9B%B4%E8%BE%BEGPU/</link>
      <pubDate>Wed, 02 Sep 2026 00:00:00 +0000</pubDate>
      <guid>https://index.zshipu.com/ai001/post/20260902/%E7%BE%8E%E5%85%89%E8%BF%91GPU-NAND%E9%97%AA%E5%AD%98%E8%AE%A9%E6%95%B0%E7%99%BEGB%E5%AD%98%E5%82%A8%E7%9B%B4%E8%BE%BEGPU/</guid>
      <description>美光正研究高耐久性NAND闪存模块，计划部署在距离GPU更近的位置，而非通过多协议连接远端存储池。这一方案允许GPU直接访问数百GB NAND存储池，用于处理非低延迟高带宽工作负载。 美光把NAND直接拉近GPU的部署逻辑 传统AI训练和推理系统中，NAND闪存通常位于远端存储池，通过</description>
    </item>
  </channel>
</rss>
