<?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>GPU调度 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/gpu%E8%B0%83%E5%BA%A6/</link><description>Recent content in GPU调度 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Mon, 13 Jul 2026 02:20:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/gpu%E8%B0%83%E5%BA%A6/index.xml" rel="self" type="application/rss+xml"/><item><title>大模型推理服务的负载均衡策略：从轮询到智能调度</title><link>https://guijiagi.com/posts/article-57/</link><pubDate>Mon, 13 Jul 2026 02:20:00 +0800</pubDate><guid>https://guijiagi.com/posts/article-57/</guid><description>深入探讨大模型推理服务中的负载均衡策略，涵盖传统算法局限、请求感知调度、模型路由等前沿方案</description></item><item><title>Agent可扩展性设计：从单机到K8s集群</title><link>https://guijiagi.com/posts/agent-scalability-single-to-k8s/</link><pubDate>Tue, 30 Jun 2026 10:20:00 +0800</pubDate><guid>https://guijiagi.com/posts/agent-scalability-single-to-k8s/</guid><description>全面解析Agent系统从单机部署到K8s集群的扩展路径，涵盖水平扩展、垂直扩展、GPU调度、自动伸缩等关键实践</description></item></channel></rss>