- What Cluster Mesh is and how it connects clusters.
- How to enable cross-cluster connectivity so pods and applications can communicate across clusters.
- How to load balance traffic across multiple—or all—clusters.
- How to configure inter-cluster network policies to control which clusters (or pods from which clusters) are allowed to talk to which other clusters.

- Full-mesh connectivity: every pod in every cluster can reach every other pod.
- Hub-and-spoke (or federated) topologies: a subset of clusters act as aggregation points.
- Selective connectivity: only specific clusters, namespaces, or services are permitted to communicate.
- Geo-aware load distribution: direct traffic to nearest or healthiest cluster.
Cluster Mesh lets you treat multiple clusters as a cohesive network domain while still enforcing cluster-local policies. Use selective connectivity to reduce attack surface and control egress/ingress paths between clusters.
| Topic | Purpose | Practical example |
|---|---|---|
| Cross-cluster connectivity | Enable pod-to-pod and service-to-service communication | Expose a central database in Cluster A to app pods in Cluster B using secure tunnels |
| Load balancing across clusters | Distribute traffic across clusters for resilience and locality | Route user requests to the nearest healthy cluster or balance traffic evenly across all clusters |
| Inter-cluster network policies | Enforce which clusters/namespaces/services may talk to each other | Restrict a sensitive service so only pods from Cluster X can connect |
Always enforce least privilege when authorizing inter-cluster traffic. Combine network policies with identity and TLS controls to limit which workloads can communicate across clusters.