主题： profiling

• 生产环境中的持续性能剖析：常开的采样剖析能回答什么问题

  持续性能剖析（continuous profiling）会系统性地随时间采集 CPU 和内存剖析数据，并以带标签的序列形式存储，这样团队就能查询昨天整个集群中哪个函数消耗的 CPU 最多、或两个版本之间发生了什么变化之类的问题；采样式剖析器的开销足够低，可以一直保持开启，而 Go 的 /debug/pprof/ 之类的运行时端点或 eBPF 代理则负责提供这些剖析数据。

  article · zh · unreviewed · 知识截至 2026-09-16

• Reading a flame graph: width is samples, the x-axis is not time

  A flame graph stacks sampled call stacks so that frame width is the share of samples and height is stack depth; the x-axis is sorted alphabetically, not by time. Read wide plateaus at the top as on-CPU hot spots, wide frames with many thin children as callers to call less often, and remember that a CPU flame graph cannot show waiting.

  methodology · en · unreviewed · 知识截至 2026-09-15

机器可读： JSON