I did wrong with 3390-6, in fact it was a 3390-9 clone. I don't have
problems with the track size, I have a problem with the maximum number of
tracks for datasets. In general, when one want to store a dataset on disk
(backup, archive) via adsm, adsm looks if there is a disk volume available
where the dataset fits into (if the file is smaller than maxfilesize). If
not, the dataset goes to tape (or whatever is next in stgp hierarchy).

My problem is to store larger files on disk to limit tape drive usage.
Today I use volumes with 2.3G. To back up files with 1 GB size, my migr.
thresholds are 40/5.  So this means half of my disk pool is empty, always.
With a single volume of 12 GB, thresholds of 75/50 may be ok, my 1Gig file
will fit, and larger files will fit,too, and I can use less dasd space.

If the usage of tape drives increases to limit because of a full disk pool,
I have lost. Migration has to wait for backups to end, so the disk pool
becomes definitively full and all further backup goes to tape.

As far as I know, I can't use multivolume datasets with adsm, but I can't
find it in the manuals.

Regards

Roger