Avamar Replication is the process of logically copying data from one or more source Avamar servers to a destination or target Avamar server.

By logically copying the data from the Avamar source to the target we are sending the backup data but not the actual data containers (also known as 'stripes') which the data reside in on the source machine.

Essentially, when the data arrives at the target Avamar server, the target will add the data to pre-existing containers (stripes) or create new ones.

Remember that the two Avamar servers are independent and not mirrors.  They function individual systems and therefore maintain their own indices and internal numbering of the data containers.

If we replicate data from, say, container 0.0-309 on the source, the target might decide to write that data to its own container ID 0.1-592 or create a brand new container altogether.

Replication is independent of that lower level functionality.

You could loosely compare this to the difference between mirroring data to a second disk in a RAID-1 array or copying the same data across the network to a CIFS share on another computer.  The underlying physical constructs are not copied over whereas the logical data is.