If I have understood correctly the objective here is to be able to
keep different projects' data separate when the backups migrate to
tape.  The problem is that with all data from all projects being
backed up to one disk storage pool it then becomes impossible to
determine which files should migrate to which tape storage pool.

I don't currently have an ADSM system to test this on but nowhere in
the documentation I have does the syntax of the "define stgpool"
command actually specify the size of the storage pool; it only sets a
limit on the maximum size of any file which is to be backed up into
it.

The actual constraint on the size of data you can write is the sum of
the capacity of the ADSM "volumes" which belong to the storage pool.
 Can a volume belong to more than one pool?  If so the solution may be
to define your disk as one big ADSM volume, assign it to a whole bunch
of storage pools and then set your management classes with appropriate
include/exclude lists.  Thus project 1's backups go to stgpool prj1
(allocated on volume adsmvolx) which migrates to stgpool prj1tape;
project 2 goes to stgpool prj2 (also allocated on volume adsmvolx) and
migrates to stgpool prj2tape etc.  When project 1 has a lot of data to
backup stgpool prj1 uses a lot of its volume but when there is not
much data the space on adsmvolx is available for other stgpools.

I have an awful feeling that a volume cannot be shared between
stgpools but this would seem to be an appropriate solution.  Anyone
game to give it a try or can ADSM Development offer an alternative?

Murray Nicholas
IT Systems Consultant
Galeforce Information Services Pty Ltd
email:  murray AT galeforce.bu.aust DOT com