Limits, Requests, QoS... Rethinking What You Thought You Knew About Scheduling in Kubernetes
summary
Explains how Kubernetes scheduling and resource controls work by examining Limits, Requests, QoS classes, and PriorityClasses. Covers what happens when cluster CPU is exhausted, how pods can exceed allotted RAM, why CPU limits can harm frontend performance, and how the scheduler decides placement — illustrated with live demos that stress clusters.
AI-written summary · about
chapters (13)
- 0:00Kubernetes scheduling intro
- 3:00Scheduler basics
- 4:05Limits and requests
- 8:59CPU limits demo
- 10:23CPU limits impact
- 18:07Memory limits
- 20:12Requests in scheduling
- 21:56Scheduling demo with Quok
- 26:10CPU shares and requests
- 27:35Cluster full behavior
- 31:48QoS pod eviction
- 33:54Priority classes
- 36:25Summary
related talks
45mLimits, Requests, QoS, PriorityClasses: Demystifying Scheduling in K8s!
21mDemystifying Resource Management in Kubernetes
24mResource Requests and Limits Under the Hood: The Journey of a Pod Spec
34mKubernetes Scheduling Features or How Can I Make the System Do What I Want?
43mEverything You Ever Wanted to Know About Resource Scheduling, But Were Afraid to Ask
36mCapacity Scheduling for Elastic Resource Sharing in Kubernetes
34mRunning prioritised workload in Kubernetes
47mOptimizing Kubernetes Resource Requests/Limits for Cost-Efficiency and Latency