<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Continuous Batching on Hellokitty&#39;s Blog</title>
    <link>https://yangyang233333.github.io/tags/continuous-batching/</link>
    <description>Recent content in Continuous Batching on Hellokitty&#39;s Blog</description>
    <generator>Hugo</generator>
    <language>zh-CN</language>
    <lastBuildDate>Tue, 25 Aug 2026 20:40:00 +0800</lastBuildDate>
    <atom:link href="https://yangyang233333.github.io/tags/continuous-batching/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Mini-SGLang 源码阅读（二）：Scheduler、Continuous Batching 与 Chunked Prefill</title>
      <link>https://yangyang233333.github.io/posts/mini-sglang-source-reading-scheduler/</link>
      <pubDate>Tue, 25 Aug 2026 20:40:00 +0800</pubDate>
      <guid>https://yangyang233333.github.io/posts/mini-sglang-source-reading-scheduler/</guid>
      <description>先解释在线推理调度的资源约束，再分析 Mini-SGLang 如何组织 Prefill、Decode 和 CPU/GPU 重叠执行。</description>
    </item>
  </channel>
</rss>
