We run an AIX ADSM server and currently have 4 storage pools, which we
would like to merge to 1 storage pool.  I'm unsure as to the procedure
to do this, given the underlying AIX structure.  Here's the current
environment:

AIX
Logical Volume
Groups                   ADSMVG1                   ADSMVG2
                              8 physical disks            8 physical
disks (a different 8)

AIX Logical
Volumes                 LV1L01 (DB vol 1)         LV2L01 (DB mirror vol
1)
                              LV1L02 (DB vol 2)         LV2L02 (DB
                              LV1L02 (DB vol 2)         LV2L02 (DB
mirror vol 2)
                              LV1L03 (/DISK01)         LV2L03 (/DISK02)
                              LV1L04 (/FILE01)          LV2L04 (/FILE02)
                              LV1L06 (Log vol)           LV2L06 (Log
mirror vol)

The /DISK01, /FILE01 pools are striped across the 8 physical disks in
VG1.
The /DISK02, /FILE02 pools are striped across the 8 physical disks in
VG2.

What we want to do is to combine all 4 pools so that we just have a
DISK01 pool.

1.  We'd like to do this without redefining the AIX logical volume
groups.
     a)  Is this possible?  That is, can an ADSM storage pool span
multiple
          AIX logical volume groups?   We'd like the DISK01pool to have
half
         of it's volumes in ADSMVG1 and the other half of it's volumes
in
         ADSMVG2.
     b)  If this is possible, how do we do this (i.e. what are the AIX
and ADSM
          steps)?  We don't have any 'unallocated' space to create
'copies' of
          our pools in the merge process.
          If we try to make the changes from ADSM only (i.e. by deleting
the
          volumes from DISK02 and then adding them to the DISK01 pool),
          aren't we now only using 1/2 of our disks (i.e. since we are
adding
          storage volumes to /DISK01 and /DISK01 is in VG1 which is on
          8 physical disks?

2.  If we can't have 1 ADSM Storage Pool span multiple logical volume
groups,
     then we are thinking of creating the following environment:
     VG2 -  3 disks (to contain the primary db and log volumes)
     VG1 - 13 disks
             - to contain the db and log mirrors
             - to contain the DISK01 pool (/DISK01 file system).
             - /DISK01 will be striped across all 13 volumes.

     The problem I see with this option is if we stripe our /DISK01
     pool across all 13 disks, if we lose a physical disk (we don't have
RAID,
     so this could happen), then we lose use of this whole pool, and all
our
     backups are out of the water.  Is this true?  If so, is there a way
     to reduce this exposure?  (today if we lose a physical disk, we
only
     affect 1/2 of our storage pools -- we can still continue 1/2 of our
backups
     to the pools in the other volume group).

Any AIX experts out there willing to take a crack at this?  Thanks.

(signed:  Hopelessly a mainframe HSMer trying to learn Unix...)