Use Karpenter to scale, optimize & upgrade K8s
summary
Karpenter replaces node pools and autoscaling groups and calls the EC2 API directly, cutting typical scale-up latency to about a minute in the demo (five pending pods scheduled to new nodes). It picks cheapest instance types including Graviton and spot, enforces per-node-pool CPU limits (demo hit a 10-CPU cap then raised to 1,000), performs consolidation when nodes are empty or underutilized (~50% default), and supports automatic node upgrades and exportable metrics for Prometheus observability.
AI-written summary · about
chapters (9)
- 0:00Introduction and Audience Engagement
- 0:12Understanding Kubernetes Challenges
- 0:51Introduction to Carpenter
- 1:27Challenges with EKS and Kubernetes
- 3:25How Carpenter Solves These Challenges
- 4:36Key Features and Configuration of Carpenter
- 8:35Live Demo of Carpenter
- 10:11Advanced Features and Cost Optimization
- 18:27Q&A Session
related talks
27mHarness the power of Karpenter to scale, optimize & upgrade K8s
22mKarpenter Efficient scaling of Kubernetes clusters
19mJust in time Worker Nodes with Karpenter
36mGroupless Autoscaling with Karpenter
18mKarpenter : Your trusted Companion for Kubernetes Scaling
01h 01mJust-in-time Nodes for Any AWS EKS Cluster: Karpenter
34mReduce Costs and Optimize on Your Cloud with Karpenter
31mKubernetes workloads with Karpenter & Amazon EKS