<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>MASH Sort - Adaptive GPU&#43;CPU Sorting on Voxell</title>
    <link>https://voxell.ai/mash/</link>
    <description>Recent content in MASH Sort - Adaptive GPU&#43;CPU Sorting on Voxell</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 01 Jan 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://voxell.ai/mash/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Sorting on Blackwell: 9× Faster Where It Actually Matters</title>
      <link>https://voxell.ai/mash/sorting-on-blackwell/</link>
      <pubDate>Thu, 01 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://voxell.ai/mash/sorting-on-blackwell/</guid>
      <description>&lt;p&gt;If your data is already almost sorted, why are you paying to sort it again?&lt;/p&gt;&#xA;&lt;p&gt;For more than a decade, NVIDIA’s CUB DeviceRadixSort has been the default answer to that question: fast, battle tested, and completely indifferent to the shape of your data. It treats a perfectly monotonic HFT order book the same way it treats cryptographic white noise. Your latency budget and your power bill pay for that indifference.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
