<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Systems on Edwardesire</title><link>/tags/ai-systems/</link><description>Recent content in AI Systems on Edwardesire</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>&amp;copy; 2021 &lt;a href="https://www.edwardesire.com/">Edward Desire&lt;/a></copyright><lastBuildDate>Tue, 22 Sep 2026 23:32:03 +0800</lastBuildDate><atom:link href="/tags/ai-systems/index.xml" rel="self" type="application/rss+xml"/><item><title>KubeCon China 2026技术观察：AI系统的资源调度与复用</title><link>/posts/kubecon-china-2026-ai-systems-resource-scheduling-and-reuse/</link><pubDate>Tue, 22 Sep 2026 23:32:03 +0800</pubDate><guid>/posts/kubecon-china-2026-ai-systems-resource-scheduling-and-reuse/</guid><description>在大规模AI集群中，提高资源利用率的手段通常从资源空闲情况开始：一张GPU还有多少显存和算力可以分给其他任务？ 当在线推理服务和Agent应用逐渐增多，</description></item></channel></rss>