How does Volume Replication work?

Volume replication is very similar to RAID 1.

The first stage involves rebuilding the destination as a copy of the source volume.  The next stage is a continuous online modification of the destination volume to preserve the mirror copy. The difference between this process and RAID 1 is that, if a connection is broken, only the changed data will be 'rebuilt', (going back to Stage 1), on the destination.

NOTE:  The replication task should be performed on a separate NIC and the connection speed should be as fast as the main network connection, to prevent lowering data throughput.

Once you create the Target Volume, you can choose to set it as a Source or a Destination in relationship to the mirrored server.  The replicated volumes need to be exactly the same size.