<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Stargz on *scratch*</title>
    <link>https://www.scrivano.org/tags/stargz/</link>
    <description>Recent content in Stargz on *scratch*</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sat, 06 Jun 2026 10:03:54 +0000</lastBuildDate>
    <atom:link href="https://www.scrivano.org/tags/stargz/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Run containers without pulling images</title>
      <link>https://www.scrivano.org/2019/10/24/run-containers-without-pulling-images/</link>
      <pubDate>Thu, 24 Oct 2019 18:37:23 +0000</pubDate>
      <guid>https://www.scrivano.org/2019/10/24/run-containers-without-pulling-images/</guid>
      <description>&lt;p&gt;CRFS is a Google project that aims at running a container without pre-pulling the image first. The key insight is that in practice a container process only accesses a small fraction of the files in its image, so fetching the entire image before startup wastes both time and disk space. CRFS achieves this through the stargz (Seekable tar.gz) format, which restructures each compressed layer so that individual files can be fetched on demand rather than requiring the entire tarball to be downloaded and extracted upfront.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
