Service and endpoints have been core primitives in K8s since 1.0. As k8s deployments gets larger, the scalability of k8s service endpoints has become a bottleneck, causing high overhead on network, cpu and memory across the cluster. In addition, the size of the endpoints object will be larger than what can be stored in the Etcd database. This talk will deep dive into the internals of k8s service endpoints API and sort out the associated scalability challenges in the current design. Minhan and Wojtek will introduce a new K8s endpoints API design that aims to improve Endpoint scalability by 100x, as well as an overview of the ongoing effort to revamp K8s Service.
Wojciech is working on Google Technical Infrastructure & Cloud since 2012. Since 2015 he works on Kubernetes and GKE. With the main focus on scalability, performance and reliability, he gained experience and contributed to many Kubernetes features and most of its components. Before... Read More →
Minhan Xia has been a member of Kubernetes networking team at Google since K8s 1.0 2015. He has contributed to various aspects of K8s networking, including pod networking, K8s service and K8s ingress.
