How We Manage Traffic for Thousands of Pods in a Multicluster Kubernetes Environment
summary
Describes how to provide transparent pod-to-pod networking across many Kubernetes clusters so they behave like a single large “supercluster.” The solution uses Istio for service mesh functionality but explicitly does not use Istio Multicluster; the talk explains the authors’ approach to routing and traffic management between thousands of pods in a multi-cluster environment.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:09Why multicluster in Kubernetes
- 3:15Traffic management problems
- 5:03Alternative solutions
- 7:07Service Mesh choice
- 9:29Solution requirements
- 12:38DNS and traffic localization
- 14:23Routing operator
- 16:52Discovery optimization
- 18:57Discovery high availability
- 20:43Summary and security
- 23:28Solution scalability
- 25:15Future plans
- 26:59Q&A
related talks
24mNavigating the Maze of Multi-Cluster Istio: Lessons Learned at Scale
23mMaking thousands of ingresses available on independent K8s clusters
06mKeynote: From One to Many, the Road to Multicluster
37mUsing Istio to Manage the Cross-Regional & Cross-Cluster...
50mMulti-Cluster Kubernetes and Service Mesh Patterns
34mIstio Multi-Cluster Service Mesh Patterns Explained
34mSpreading Apps, Controlling Traffic & Optimising Costs in Kubernetes
12mService Mesh & Kubernetes Multicluster Communication With Linkerd