Local area replication (LAR) allows logical separations to be created in a storage cluster to define storage distribution strategies. These logical separations cause Swarm to attempt to create the greatest logical spread between an object's replicas by moving them into separate subclusters. Examples where LAR subclusters are useful include:

• Splitting a cluster based on location (data cabinet, building wing)

• Grouping nodes based on common infrastructure (network, power)

Consider splitting a cluster based on location if data centers are located in separate wings of a building and copies of stored content are desired to exist in both locations in case of a partial building loss. Grouping the cluster nodes based on a common infrastructure can be another option if you want to group cluster nodes by shared network switches, a common power distribution unit (PDU), or a common electrical circuit within a rack.

Network Requirements

The network connections between LAR subclusters must have the same speed and latency characteristics as the connections between the nodes. All nodes must be in the same broadcast domain, so they are able to send data directly to all nodes in the cluster and receive the multicast traffic sent from anywhere in the cluster.

Space Requirements

Verify sufficient space exists in the LAR subcluster containing the retiring volumes if the separation to persist when retiring a volume is desired. Retiring a volume without sufficient space can be problematic because Swarm must maintain the correct number of replicas in the subcluster. Swarm may create all replicas on the other side of the subcluster, filling up that side of the subcluster.

See node.subcluster in the Settings Reference.