Hi,
We run ADSM 3.1 on AIX with a StorageTek DLT cabinet
w/6 drives.
We're starting to run into some minor drive
availability issues.  I thought of a potential
solution (it should at least help), but don't know how
feasible it is.  Here's the situation:

Our backups go to disk.

During the day, we migrate the data from disk to DLT
tape using
   update stgpool backuppool
where "stgpool" is the onsite DLT pool, and
"backuppool" is the disk

Then, we copy the data from the onsite DLT pool to the
offsite DLT pool using
   backup stg dltops dltoffsite
where "dltops" is one of our onsite DLT pools, and
"dltoffsite" is the offsite DLT pool.

So the data trail is:
Disk->OnsiteDLT then OnsiteDLT->OffsiteDLT (which
requires 3 drive mounts in a 24 hour period)

Because we've got a handfull of storage pools to copy
to "dltoffsite", and a few other processes which go
directly to tape, we've been running into drive
availability issues.

My question is this, is there some way to make, both,
onsite and offsite copies straight from disk to tape?


The new data trail would be:
Disk->OnsiteDLT then Disk->OffsiteDLT (which requires
2 drive mounts in a 24 hour period)

Beside using up fewer drives, it seems that the
migration from disk to tape is considerably faster
than the copy from tape to tape, e.g. one of our
migrations is set to run 4 hours.  The tape to tape
copy can sometimes take >12.

The one reason I can see that this would not work is
in the case of restoring volumes.  Does ADSM build the
relational database between onsite and offsite during
our copy, making the copy process necessary? or can it
build the database purely by matching the data between
onsite and offsite volumes?

Thanks,
ch

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail - only $35
a year!  http://personal.mail.yahoo.com/