Skip to content
Chaos Lab

Challenge 36 · starter · Pricing API

CPU bottleneck

The pricing API runs on 2 instances with 4 cores each and normally takes 1,200 requests a second. At 10 seconds the lunchtime peak doubles traffic. Someone left the log level at debug after an incident last week.

From 15 to 60 seconds, answer at least 99.5% of requests with p99 at or under 150 ms, on no more than 2 instances.

ready00:00 / 01:00Success · 10 s–p99 · 10 s–

Goal · measured 00:15–01:00

0 changes · par 1
  • Requests answered ≥ 99.5%not measured yet
  • p99 latency ≤ 150 msnot measured yet
  • Instances ≤ 2 instancesnot measured yet

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.