Demo: latency caused by Kubernetes CPU throttling
On This Lesson 3 topics
This lesson is planned. The outline below lists the topics it will cover. The full lesson has not been published yet.
In this lesson
find a cross-layer cause where the app looks fine but the container is throttled; prepares SRC-015-LAB1.
Draft outline details
Step by step
- Symptom: p99 spikes in bursts while average CPU looks low.
- Rule out the database and GC from their metrics.
- Check container metrics for CPU throttling (cgroup throttled periods); correlate with latency spikes.
- Planned mistake: raise the JVM heap. Fix: the evidence points at CPU limits, so adjust the limit or request, and review JVM thread and GC thread counts relative to the container CPU; verify JVM container awareness for your version.
- Re-test and confirm the spikes disappear.
Check your understanding
Why can average CPU look low while throttling harms latency?
Show answer
Throttling occurs in short bursts when the quota is used up within a period.
Practice
Apply this lesson with the module's lab, exercises and quiz.
Marking a lesson complete is optional and you can undo it any time. Your progress stays in this browser.