Defusing the Kubernetes API Performance Minefield - Madhav Jivrajani & Marek Siarkowicz
Madhav Jivrajani, Marek Siarkowicz
KubeCon + CloudNativeCon Europe 2025 · Session
This talk, presented by Marek Siarkowicz, SIG lead of API Machinery and contributor to etcd, delves into the critical and often overlooked realm of **Kubernetes API performance**. While Kubernetes is celebrated for its ability to manage containerized workloads at scale, the underlying API server, particularly when interacting with Custom Resource Definitions (CRDs) and operators, harbors significant performance pitfalls. Madhav Jivrajani, unfortunately unable to attend, was also a key contributor to this work. The presentation aims to expose these "minefields," explain their root causes, and present recent and upcoming solutions designed to make the Kubernetes API server more robust and efficient.
AI review
This talk from Marek Siarkowicz provides a brutally honest and deeply technical exposition of critical Kubernetes API server performance issues, specifically "minefields" related to CRDs and large list operations. It meticulously details the "five allocation" problem causing catastrophic memory spikes and then presents the elegant, core architectural solutions: Consistent Reads from Cache (K8s 1.31), List from Snapshot with b-trees (K8s 1.33 Alpha), and Streaming Collection Encoding (K8s 1.33 Beta). The presented benchmarks, showing a reduction from 70GB to 3GB memory usage for list…