Local area replication (LAR) allows you to create logical separations 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)
For example, if you have data centers located in separate wings of your building and you want to have copies of stored content exist in both locations in case of a partial building loss, consider splitting your cluster based on location. However, if you want to group your cluster nodes by shared network switches, a common power distribution unit (PDU), or a common electrical circuit within a rack, grouping the cluster nodes based on a common infrastructure can be another option.
Note
Network requirements. The network connections between LAR subclusters must have the same speed and latency characteristics as the connections between the nodes. Additionally, all nodes must be in the same broadcast domain such that they are able to send data directly to all other nodes in the cluster and receive the multicast traffic sent from anywhere in the cluster.
Warning
Space requirements. When you retire a volume, make sure that sufficient space exists in the LAR subcluster that contains the retiring volumes if you want the separation to persist. Because Swarm must maintain the correct number of replicas in the subcluster, retiring a volume without sufficient space can be problematic. For example, Swarm might create all replicas on the other side of the subcluster, filling up that side of the subcluster.
See node.subcluster in the Settings Reference.