Challenge 19 · intermediate · CPU limits
CPU throttling
A search service runs as one pod with a 2-core CPU limit and 16 worker threads, on a node with 8 cores. Each request needs about 40 ms of CPU. At 40 requests a second the dashboard says the pod uses about 1.7 cores, comfortably under its limit, yet some searches take half a second. At 20 seconds traffic rises by half.
From 5 to 80 seconds, answer at least 99% of requests, with a p99 of 200 ms or less.
Goal · measured 00:05–01:20
0 changes · par 1- Requests answered ≥ 99.0%not measured yet
- p99 latency ≤ 200 msnot measured yet
Service map
What changed
Turn on a fault or change a defence while the run plays. Its effect shows here: the 5 seconds before against 5 seconds after it settles.
This is a simulation. Its numbers come from a model of a typical system, not from measuring real servers. The same link and seed always give the same run. How the Chaos Lab works.