<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>高性能网络 on PlumePHP</title><link>https://plumephp.com/tags/%E9%AB%98%E6%80%A7%E8%83%BD%E7%BD%91%E7%BB%9C/</link><description>Recent content in 高性能网络 on PlumePHP</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Sat, 26 Sep 2026 10:30:00 +0800</lastBuildDate><atom:link href="https://plumephp.com/tags/%E9%AB%98%E6%80%A7%E8%83%BD%E7%BD%91%E7%BB%9C/index.xml" rel="self" type="application/rss+xml"/><item><title>事件驱动 IO 与高性能网络模型：epoll/Reactor/io_uring</title><link>https://plumephp.com/network-io-event-driven/</link><pubDate>Sat, 26 Sep 2026 10:30:00 +0800</pubDate><guid>https://plumephp.com/network-io-event-driven/</guid><description>&lt;p&gt;高并发服务端的本质是把「等待」交给操作系统，把「计算」留给 CPU。从每线程一个连接的阻塞模型，到 epoll 驱动的 Reactor 模型，再到 io_uring 的全面异步，IO 模型的每一次进化都在提升单机并发上限。这篇文章系统梳理这条演进主线。&lt;/p&gt;</description></item></channel></rss>