But this is only a factor when using the simultaneous write capability
which isn't the default (and in my experience isn't used all that
frequently). In more common usage a backup storagepool process will,
once invoked, read uncopied data from primary storagepool volumes and
write it to copy storagepool volumes - if you've significantly slower
disk for your copy storagepool volumes (which, as Richard points out,
would have to be configured as sequential volumes) then I'd suggest a
bottleneck could probably be the write throughput on these disks.

/David Mc
London (and off to Dubai next week!)

On 16 Jan 2010, at 10:50, Mehdi Salehi <iranian.aix.support AT GMAIL DOT COM>
wrote:

From TSM Concepts Redbook:
"Because the data is written to the copy storage pool and primary
storage
pool simultaneously, the backup performance will only be as good as
the
slowest device being used for any of the pools."