<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Mimalloc on Luther Monson</title>
    <link>https://luth.io/tags/mimalloc/</link>
    <description>Recent content in Mimalloc on Luther Monson</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Thu, 15 Oct 2026 19:45:00 -0700</lastBuildDate>
    <atom:link href="https://luth.io/tags/mimalloc/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Fourth Place in Worker Mode, and Why It Wasn&#39;t PHP</title>
      <link>https://luth.io/blog/2026/10/ephpm-worker-mode-performance/</link>
      <pubDate>Thu, 15 Oct 2026 19:45:00 -0700</pubDate>
      <guid>https://luth.io/blog/2026/10/ephpm-worker-mode-performance/</guid>
      <description>&lt;p&gt;We added &lt;a href=&#34;https://github.com/ephpm/ephpm&#34;&gt;ePHPm&lt;/a&gt; to the&#xA;&lt;a href=&#34;https://github.com/Yii3-Benchmarks/app-api&#34;&gt;Yii3 runtime benchmark&lt;/a&gt;. It runs the same Yii3 API app on a bunch of PHP&#xA;runtimes and compares them. Our fork with the ePHPm runner is &lt;a href=&#34;https://github.com/ephpm/app-api&#34;&gt;ephpm/app-api&lt;/a&gt;.&lt;/p&gt;&#xA;&lt;p&gt;Most of the results were good. ePHPm came first in classic mode and first on the database endpoint in both modes.&#xA;Classic mode is the PHP-FPM style, where every request starts fresh.&lt;/p&gt;&#xA;&lt;p&gt;Worker mode was the surprise. In worker mode the app boots once and stays in memory, like Octane or FrankenPHP&amp;rsquo;s&#xA;worker mode. We came fourth of ten at about 50.7k requests a second. Workerman (~65k) and Swoole (~62k) were ahead.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
