Chris,

I guess I could have described the problem a little better.  What I have is
SSA disks backed up by a 3583 LTO library w/6 drives?  And this works very
well.  It's when I get this DB2 backup once a week that bothers me.  What I
was doing was setting the POLICY so that all data from several UNIX systems
went directly to tape when I knew the DB2 Admin was going to do the backup.
Last night I forgot to do that and I was worried that she might lock the
system because the 125GB wouldn't fit into the backup disk storage pool.
I guess it worked because I didn't get any calls and after quizzing her
this morning she was happy.

So what I'm thinking is that TSM looked at the size--not sure how it knew
the size unless the DB2 api does something--and it bypassed the disk pool
and went directly to tape.

Sam






Chris Gibes <gibes AT BERBEE DOT COM> on 06/21/2001 02:19:07 PM

Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>



 To:      ADSM-L AT VM.MARIST DOT EDU

 cc:      (bcc: Sam Schrage/Nashville)



 Subject: Re: Disk pool size vs large file









A couple of questions first. What kind and how many tape drives are you
migrating to? What kind of disk are you using and how is it attached? I
have a seen a few sites with SSA disk pools and multiple 3590 tape drives
with a backups over a 100meg ethernet connection where migration from disk
to tape would happen at the same time as backups to disk with no problems
at all.  Even with multiple migrations over the course of a backup. The key
point is whether the process of moving data from the diskpool to tape is
faster than data moving from the client node to the diskpool. If disk -->
tape is faster than client --> disk then I don't see a huge problem.

Most of the time in this type of situation the bottleneck is the network
not the disk or the tape.

As always ymmv.

At 01:14 PM 6/21/01 -0500, you wrote:
>I have a 56GB disk pool with the next pool to tape.  I have a user, DB2
>Admin, that wants to back up a 125GB DB2 backup.  What's the best way to
>handle this one user?  If he/she backs up the file will it crash the
system
>because it's bigger that the diskpool, or will it go right to tape?
>
>Sam Schrage
>TRW Systems
>sam.schrage AT trw DOT com

Chris Gibes <gibes AT berbee DOT com>
Berbee- Milwaukee Branch
N14 W23822 Stone Ridge Dr.
Waukesha, WI 53188
(262) 521-5616


Berbee...putting the E in business