Customer has NSBS66 SP4a 2 node cluster with 3rd server hosting iSCSI
shared storage. Shared storage server is also Backup Exec 9.1 media
server with HP Ultrium 1 drive. No backup software is installed on the
cluster servers (other than the TSAs).

Agents are configured in BE for each cluster volume pointing to the IP
address of the resource. The backup jobs are configured to back files
up via these agents.

The problem is that backups are breathtakingly slow; about 30 - 35MB/min
is achieved. This is a problem because once all the GW data moved from
the old server to the new one, there will not be sufficient hours in a
day to back it all up.

We have used TSATEST which confirms the effective backup rate, so the
problem is the network and not the backup sub-system.

Can anyone provide any pointers as to how backup performance can be
improved in this scenario?