<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Silicon &amp; Systems (EN)</title>
    <link>https://siliconandsystems.com/en/</link>
    <atom:link href="https://siliconandsystems.com/en/rss.xml" rel="self" type="application/rss+xml"/>
    <description>An independent tech review covering semiconductors, computer architecture, systems, and AI infrastructure. Slower than news, easier than papers.</description>
    <language>en</language>
    <item>
      <title>Why AI Datacenters Want to Become One Chip</title>
      <link>https://siliconandsystems.com/en/articles/cxl-onechip/</link>
      <guid isPermaLink="true">https://siliconandsystems.com/en/articles/cxl-onechip/</guid>
      <pubDate>Mon, 24 Aug 2026 00:00:00 GMT</pubDate>
      <description>A training step completes only after every participant reaches the same barrier. A recent review in Nature Reviews Electrical Engineering argues that addressing this at rack scale is a silicon problem rather than a bandwidth problem, and identifies the three hardware structures it requires.</description>
    </item>
  </channel>
</rss>
