I don't think you can find out exactly; but the amount of DB space is
proportional to the number of objects backed up (or archived).

You can get that with a select on the backups table.

Now you have to be careful with selects on the backups table on a large DB;
my DB is 33 GB and a select over the whole backups table just never
completes.

BUT, that table is indexed by node_name.

What I have done in the past is to write a host script that pulls off a list
of the node names:
select node_name from nodes where domain_name='your.server.domain' (or
however else you can identify those machines)

Then cycle through each node name with
select count(*) from backups where node_name='BLAH'

Repeat for archives, if you have a lot of archives.
That works for me.




-----Original Message-----
From: MC Matt Cooper (2838) [mailto:Matt.Cooper AT AMGREETINGS DOT COM]
Sent: Friday, September 05, 2003 9:27 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: how much of the DB does a node use?


Hello all,
        I am trying to group the 200 server + 300 PC nodes that I backup
into meaningful groups so I can seperate my TSM server into 3 TSM servers.
I am currently running TSM 5.1.5.4 on a z/OS 1.1 system.  I have to keep all
processing off of 1st shift and am thinking I can increase my speed by using
more processors and cutting down on the internal TSM conflicts (making
offsite copy while doing backups or expiration and migration together , etc)
by splitting my TSM server into 3 TSM servers.   I have enough excess
processor during off hours and management is willing to give me more disk
and tape drives (within reason) to make this work.   I can easily break off
the PCs and all Development clients into a TSM instance.  They can be backed
up late on 3rd shift and use the available tape drives on early 2nd shift.
The SLA for that group allows the offsite copy to be a day late.   But when
I get to the group of true servers that TSM backs up I want to know the
impact on the TSM DB size for my decisions as which nodes go to which TSM
server.   My server clients are pretty diverse from hug DB2 data bases that
look like half a dozen 30GB files (little TSM DB impact) to Window servers
with millions of files (high TSM DB impact).
        So the question is,  How can I find out how much of the TSM DB does
each of my client nodes use?
THanks in advance
Matt