Cluster support lets AkkEngine run with a replication runtime. It is a low-level engine feature, not a complete distributed database control plane: the application or operator still decides membership, startup role, primary target, secure identity, traffic routing, and failover procedure.
Use it when you want engine-level replication between native nodes and are comfortable managing topology outside the storage engine. If you need automatic leader election, split-brain-safe failover, online resharding, or a managed cluster API, those pieces need to be built above this layer.
Read By Goal
Section titled “Read By Goal”| Goal | Page |
|---|---|
| Configure nodes and start primary/replica processes | Cluster Usage |
| Understand runtime parts, routing, replication frames, and limitations | Cluster Architecture |
The current Cluster layer is best treated as an engine-level replication primitive. It gives you explicit placement, roles, acknowledgement policy, and replication transport, but it does not provide an external control plane.
Production deployments should add their own membership management, failover, traffic routing, snapshot transfer, health checks, and migration workflow.