I know of no TSM-based limit.  
I personally have 1672 25037MB FILE-class volumes on 88 filesystems. The
volume size was chosen to entirely fill the filesystems.  The filesystem
size was chosen to fit evenly within the LUNs.  The LUN size was chosen
to exactly fill optimally-available arrays.  The array size was chosen
to use one disk in each enclosure of the storage device.  RAID5 across
11 enclosures FAStT - 372G drives.
I can pretty much always keep the LTO3 saturated when reading from these
volumes... Though I was tempted to make one volume per filesystem, just
to keep TSM from insisting on concentrating its I/O into a single path.

-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Michael Glad
Sent: Thursday, September 18, 2008 2:15 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: [ADSM-L] Volume size on sequential disk storage pools

Hello TSM'ers, are there any recommendations/experiences concerning
volume size/ maximum number of volumes on sequential disk storage pools?

On our largest TSM server, we currently have a 27T pool consisting of
10G volumes => some 2700 volumes. I chose the 10 gig size a couple of
years ago after reading an IBM paper.

Although there currently doesn't seem to be any major performance
problems, I wonder if there's a limit to the number of volumes after
which TSM may get into troubles.

So I would be glad to learn what volume size / number of volumes other
TSM sites using sequential disk volumes use.

        -- Michael