NOTE
1.15 Inter-Node Communication
How distributed nodes exchange cluster metadata and the trade-offs between centralized coordination and peer-to-peer dissemination.
This is a historical learning note and may contain outdated or incomplete understanding.
1. Why Nodes Communicate
Cluster members exchange metadata needed to route requests and coordinate work. Examples include:
- leader/follower or primary/replica roles;
- node membership and health;
- partition ownership;
- replication progress and topology information.
2. Metadata Distribution Models
2.1 Centralized Coordination
A dedicated coordination system stores or helps agree on cluster metadata. Clients and nodes consult that system to discover ownership or leadership. ZooKeeper is a classic example of a coordination service.
The benefit is a clear authoritative coordination point; the cost is an additional critical subsystem whose availability and consistency must itself be engineered.
2.2 Peer-to-Peer
Each node owns part of the cluster view and exchanges metadata directly with peers. Redis Cluster is an example of a system that disseminates cluster information between nodes.
This removes a separate metadata service but makes convergence, conflict handling, and membership logic part of the cluster protocol.
3. Gossip
Gossip protocols repeatedly exchange membership or metadata with peers so that information eventually spreads across the cluster. See Gossip.