Taking Care of Your Control Plane With API Priority and Fairness and Resource Quotas

This conference talk explores how to maintain stability in Kubernetes multi-tenant environments using API Priority and Fairness (APF) and Resource Quotas. Learn from Datadog's experience implementing these tools across hundreds of clusters and thousands of workloads. Discover how to effectively classify and throttle API Server requests with APF while using Resource Quotas to limit resource consumption per namespace. Gain practical insights into configuration strategies, common challenges and their solutions, potential drawbacks to consider, and how to apply these lessons to your own Kubernetes clusters. Essential knowledge for cluster administrators dealing with competing workloads who want to prevent individual users from overloading the Control Plane or consuming disproportionate resources.