Cutting Costs and Boosting Efficiency with Karpenter and KEDA on Kubernetes
summary
Karpenter and KEDA for Kubernetes: using Karpenter to pick instance types (including Spot and Graviton) to cut infrastructure costs, combined with KEDA to perform event-driven and metric-based autoscaling (Kafka queues, Prometheus metrics). The talk explains how to integrate these tools so clusters scale dynamically and run more efficiently in the cloud.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 2:51Optimizing Kubernetes resources
- 7:45Karpenter cost reduction
- 13:00Karpenter node provisioning
- 16:15Node consolidation policies
- 19:26KEDA pod scaling
- 23:20KEDA architecture
- 25:50KEDA zero scaling costs
- 27:37Karpenter KEDA use case
related talks
46mKarpenter * Keda: The Winning FinOps Duo
46mScaling EDA Workloads with Kubernetes, KEDA & Karpenter
42mScaling Kubernetes-based Event-driven Workloads with Keda & Karpenter
40mOptimize EKS Cost and Scaling with Karpenter
34mReduce Costs and Optimize on Your Cloud with Karpenter
33mKarpenter: Best Practices and Cost Optimization Techniques
38mScaling EDA Workloads with Kubernetes, KEDA & Karpenter
22mKarpenter Efficient scaling of Kubernetes clusters