Virtual volume copy availability

Whether a copy is available at another TS7700 Cluster depends on the copy consistency point that had been assigned to the virtual volume when it was written.

You choose whether a TS7700 Cluster is to have a copy of a virtual volume's data and when the copy will be performed. You can select from the following data copy policies:

Copy data as part of the rewind unload processing (RUN) for a volume
Important: Since immediate mode copies take place in their entirety after a volume is closed, the ability to replicate 6000 MB volumes increases the copy and Rewind-Unload (RUN) device end holding time by 50% over 4000 MB volumes. The immediate mode replication policy is not optimal for use with 6000 MB or 25000 MB volumes.
Defer copy after the RUN for the volume
Defer (time-delay) copy to reduce traffic on the grid network and reduce excessive reclamation activity on target clusters
Copy data synchronously as part of a synchronous mode copy operation
Do not copy

If the TS7700 Cluster used as the initial source for a virtual volume fails and another TS7700 Cluster has not completed a copy, then that virtual volume is not available during the outage and the mount request for the volume within the grid fails.

Parallel copies and pre-migrations for read mounts

This enhancement allows copies and pre-migrations within the grid to continue when a private mount occurs. Read mounts can occur in parallel to replication and pre-migration activity. If a write or update occurs, all pre-migration and replication activity for the same volume restarts sometime after the volume is closed and demounted.