Which redundancy model is used in CUCM clusters?

• A. No redundancy is used; all nodes are active.
• B. N+1 style redundancy using multiple call-processing nodes and automatic failover.
• C. N+2 style redundancy with double capacity.
• D. Single-node failover with manual switchover.

Answer: (B)

CUCM achieves high availability by spreading the call-processing workload across multiple nodes and providing automatic failover. In a cluster you have several Call Processing Nodes handling calls, and you also include at least one spare node so that if any CP node goes down, the remaining nodes can take over without manual intervention. This is the N+1 redundancy model: enough active processing capacity to handle the load, plus one extra node ready to assume duties automatically if something fails. The failover happens transparently, with the database kept in sync via the publisher/subscriber setup, so ongoing calls aren’t disrupted. The other options would either leave no protection, require unnecessary extra capacity (N+2), or rely on manual switchover, which would cause downtime.