[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fdWDI2aCXW9MJairtn5xddwpJOkRxA3e3I1rBTL-JnV0":3},{"item":4,"related":44},{"id":5,"type":6,"title":7,"slug":8,"summary":9,"body":10,"coverUrl":11,"productScreenshots":12,"productLinks":13,"authorName":14,"authorUrl":15,"authorSubject":16,"category":17,"tags":22,"sourceLabel":35,"sourceName":36,"sourceUrl":37,"status":38,"seoTitle":7,"seoDescription":9,"canonicalUrl":35,"isFeatured":39,"sno":40,"sortOrder":41,"publishedAt":42,"updatedAt":43,"createdAt":43},"eab786d8-a860-4488-bf9a-7fd9c55f0237","article","PSI：CPU 没跑满，系统为什么还是会卡？","linux-psi-pressure-stall-information","PSI 从任务停顿而非单纯硬件利用率衡量 CPU、内存和 I\u002FO 压力。本文介绍 some\u002Ffull 信号、阈值监控及解读时的常见误区。","一台机器的 CPU 利用率没有达到 100%，用户却觉得系统卡顿，这并不矛盾。CPU 数字描述处理器忙不忙，却不一定说明有多少任务正在等待资源。PSI（Pressure Stall Information，压力停顿信息）试图回答另一个问题：任务因为 CPU、内存或 I\u002FO 资源不足，实际停下来等了多久？\n\n可以把系统想成一个服务大厅。看窗口工作人员是否忙碌，不等于知道顾客排队等了多久；反过来，窗口没满负荷，也可能有一群顾客被某个关键步骤堵住。PSI 从任务停顿角度衡量压力，为系统管理员提供“资源瓶颈是否正在拖住工作”的信号。Linux 的 PSI 接口区分 `some` 与 `full` 等情况：前者关注一部分任务受影响，后者关注某些时间区间内全部非空闲任务都受影响。它按资源类型提供相应统计和监控能力。\n\n内存压力尤其容易被误读。即使机器尚未完全耗尽物理内存，回收页面、等待交换或争抢内存也可能让任务停顿。I\u002FO 路径上，设备队列变长或请求等待也可能拖慢工作；CPU 压力则可能体现为可运行任务排队。PSI 把不同资源造成的停顿分别呈现，帮助你把“感觉慢”转化为更具体的观察对象。\n\nPSI 的监控接口还允许应用或管理组件设置阈值并接收压力事件，可用于在机器恶化前采取动作，例如限流、调整工作负载或释放缓存。它不是自动优化器，也不会替你决定哪个任务应该让路。阈值怎么定，需要结合业务延迟目标、正常基线和负载波动；不同工作负载的合理值并不相同。\n\n读 PSI 时，最好与 CPU 利用率、内存回收、磁盘延迟、队列长度以及服务端到端延迟一起看。单一读数不能直接证明根因：高压力可能是上游流量突然增加的结果，也可能是资源配置不足，或由某个异常任务造成。还要注意采样窗口和统计范围，避免把短暂尖峰误判成长期故障。\n\nPSI 的价值是把“资源紧张”从一个模糊标签变成可量化的任务停顿。它特别适合容器、服务器和资源管理系统观察系统何时真正拖住了工作，而不是只看某个硬件仪表盘上的百分比。\n\n来源：[Linux 内核文档：Pressure Stall Information](https:\u002F\u002Fdocs.kernel.org\u002F6.10\u002Faccounting\u002Fpsi.html)","\u002Fuploads\u002F2026-10-06\u002Ff50b161d-4230-44dd-a953-1089863b2f05.jpg",[],[],"Foundit","https:\u002F\u002Ffoundit.cn","foundit-ai-editorial",{"id":18,"name":19,"slug":20,"description":21},"6179d3b6-dc34-4483-9ded-3cd9f1b37a47","科普","abbreviation","介绍各领域新兴概念",[23,27,31],{"id":24,"name":25,"slug":26},"7c76bfc2-f80f-4ee0-a95d-27bd8708b434","技术","slug",{"id":28,"name":29,"slug":30},"4c2bbea6-eab7-40a8-8447-1de478ff7749","分析","analyse",{"id":32,"name":33,"slug":34},"68cedb55-2cac-412f-8f81-fda8c7d686dd","思考","thought",null,"Linux Kernel Documentation：Pressure Stall Information","https:\u002F\u002Fdocs.kernel.org\u002F6.10\u002Faccounting\u002Fpsi.html","published",false,60,0,"2026-10-06T00:00:00.000Z","2026-10-06T07:51:23.846Z",[45,53,61],{"id":46,"type":6,"title":47,"slug":48,"summary":49,"coverUrl":50,"authorName":14,"sno":51,"publishedAt":42,"createdAt":52},"58fc0fe2-4392-402e-bd43-68da30f7b50f","RPKI 路由源验证：怎样确认一个网络前缀确实由某个 AS 宣告？","rpki-bgp-route-origin-validation","RPKI 路由源验证通过 ROA 检查前缀与起源 AS 是否匹配。本文解释 Valid、Invalid、NotFound 的含义，以及它为何不验证整条路由路径。","\u002Fuploads\u002F2026-10-06\u002F073132a3-dfee-42bd-b7f1-9f6805946e8b.jpg",40,"2026-10-06T07:51:36.206Z",{"id":54,"type":6,"title":55,"slug":56,"summary":57,"coverUrl":58,"authorName":14,"sno":59,"publishedAt":42,"createdAt":60},"d9dbb54d-46d3-4b1b-ad49-e4e049264170","Sloppy Quorum 与 Hinted Handoff：副本故障时，数据能先寄存到别处吗？","sloppy-quorum-hinted-handoff-replica-recovery","Dynamo 风格的宽松 quorum 可在首选副本不可达时暂存写入，并通过 hinted handoff 在故障恢复后补回。本文解释其可用性收益与一致性边界。","\u002Fuploads\u002F2026-10-06\u002Fc62b7ad0-304e-4a21-a245-aeee01573342.jpg",53,"2026-10-06T07:51:44.617Z",{"id":62,"type":6,"title":63,"slug":64,"summary":65,"coverUrl":66,"authorName":14,"sno":59,"publishedAt":67,"createdAt":68},"e4131e64-25af-4b75-8c90-d78212d5b7d9","BBR：网络拥塞为什么不能只看丢包？","bbr-bandwidth-round-trip-congestion-control","BBR 通过估计瓶颈带宽和往返传播时间来调节发送速率，把吞吐、延迟与在途数据纳入拥塞控制，而不是只等待丢包。","\u002Fuploads\u002F2026-09-26\u002F7a21c7b9-a2ee-4f70-b3fe-b10fca22655d.jpg","2026-09-27T00:00:00.000Z","2026-09-26T16:26:20.942Z"]