Intelligent Workload Re-shuffling to Minimize Idle Resources Across Cluster
summary
Describes a VMware approach that detects resource fragmentation in Kubernetes clusters—when node-local resources are insufficient even though aggregate capacity exists—and produces migration plans to relocate pods so waiting pods can be scheduled. Covers a lightweight periodic rebalancing mechanism that proactively reshuffles pods across nodes to avoid future fragmentation and reduce unnecessary cluster scale-ups.
AI-written summary · about
chapters (14)
- 0:00Intro
- 0:22Agenda
- 1:05The Need
- 2:09Kubernetes Schedulers
- 4:07The Problem
- 6:44Problem 1: Placement Failure
- 8:14Solution: Migrate and Place
- 9:17Placement Strategies
- 11:20Demo 1
- 14:29Problem 2: Imbalanced Placement
- 17:28Solution: Swap and Balance
- 18:12Balancing Engine
- 23:16High Level View
- 24:50Collaborating
related talks
11mHow Jenga & Tetris Can Help Run More Containers with Lesser Costs...
29mRight-Size Your Kubernetes Deployments
38mDynamic Pod Resource Boundary Adjustment in Web Scale Clusters
38mDynamic Pod Resource Boundary Adjustment in Web Scale Clusters
05mResource Roulette: Winning the Kubernetes Allocation Game
32mOptimizing Resource Usage in Kubernetes
15mDemo: Resource(Cost) optimization on planet scale Kubernetes clusters.
36mDon't Be Greedy: Rightsize Your Kubernetes Cluster With Prometheus