ADSM-L

Fw: Restore in DR site is very slow

2006-07-28 08:32:14
Subject: Fw: Restore in DR site is very slow
From: Nicholas Cassimatis <nickpc AT US.IBM DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Fri, 28 Jul 2006 08:31:39 -0400
Increase the number of processes used for the offsite copy.  With only 2
processes, you have 2 target tapes, so the data may only be on 2 tapes in
the copypool.  If you use more processes, you're more likely to have the
data spread across more tapes.

Nick Cassimatis

----- Forwarded by Nicholas Cassimatis/Raleigh/IBM on 07/28/2006 08:31 AM
-----

"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 07/28/2006
07:54:05 AM:

> Production side:
> TSM Server: 5.3.3 on AIX 5.3
> Tape Library: 3584 with 8 LTO3 drives
> DB2 database: 8.2,   1.2TB with 7 backup drives, backup time 1.2 hour
> offisite copy, 2 processes, with 4 drives.
>
> DR site:
> TSM Server: 5.3.3 on AIX 5.3
>  Tape Library: 3584 with 4 LTO3 drives
>
>
> My problem is, when I did twice restore test, the first time it used 4
> drives to do restore, used 5 hours to restore. The second test, it only
used
> 2 drives to do restore and it used 12 hours.
>
> My question is, how can I ensure the restore can constantly use 4 drives
to
> do restore?
>
> TIA.