<?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>Docker on Brandon&#39;s Technical Blog</title>
    <link>https://brandongeraci.com/tags/docker/</link>
    <description>Recent content in Docker on Brandon&#39;s Technical Blog</description>
    <generator>Hugo -- 0.146.0</generator>
    <language>en-us</language>
    <lastBuildDate>Sat, 08 Nov 2025 12:00:00 +0000</lastBuildDate>
    <atom:link href="https://brandongeraci.com/tags/docker/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The Unified Memory Revelation: Why Docker Double-Counts</title>
      <link>https://brandongeraci.com/projects/dgx-spark/03-unified-memory-revelation/</link>
      <pubDate>Sat, 08 Nov 2025 12:00:00 +0000</pubDate>
      <guid>https://brandongeraci.com/projects/dgx-spark/03-unified-memory-revelation/</guid>
      <description>The &amp;#39;aha!&amp;#39; moment: Docker&amp;#39;s cgroups were designed for discrete GPUs, but Grace Blackwell has unified memory. The result? Docker counts GPU memory twice, creating 20-30GB of phantom overhead.</description>
    </item>
    <item>
      <title>The Mystery: Don&#39;t Just Blame the Hardware</title>
      <link>https://brandongeraci.com/projects/dgx-spark/01-the-mystery/</link>
      <pubDate>Sat, 08 Nov 2025 10:00:00 +0000</pubDate>
      <guid>https://brandongeraci.com/projects/dgx-spark/01-the-mystery/</guid>
      <description>YouTube tech reviewers were quick to blame NVIDIA&amp;#39;s new DGX Spark for being slow. But was it really the hardware&amp;#39;s fault, or was I missing something deeper in the software stack?</description>
    </item>
  </channel>
</rss>
