Dynamic Pod Resource Boundary Adjustment in Web Scale Clusters
summary
Alibaba describes a Kubernetes-native approach for co-locating Pods of different QoS classes on the same node and dynamically adjusting Pod resource limits (especially under contention) to balance resource efficiency and application SLOs. They present operational practices and production results from web-scale clusters showing improvements in cluster utilization, tail latency, and TPS.
AI-written summary · about
chapters (6)
- 0:00Start
- 1:58Our Battle
- 12:41Resource Utilization Rate
- 14:38VPA or HPA
- 16:54What Should We Do?
- 25:58Palley engine框架设计
related talks
38mDynamic Pod Resource Boundary Adjustment in Web Scale Clusters
05mResource Roulette: Winning the Kubernetes Allocation Game
23mOptimizing Storage Assignment via Pod Scheduling Under Disturbance Factors
36mTaming the Beast: Advanced Resource Management With Kubernetes
32mCoordinate Workloads Colocation: QoS-Oriented Scheduling Enhancement on K8s
29mRight-Size Your Kubernetes Deployments
15mDemo: Resource(Cost) optimization on planet scale Kubernetes clusters.
29mResize Your Pods w/o Disruptions aka How to Have a Cake and Eat a Cake