=> On Thu, 2 Sep 1999 10:17:08 -0400, "Viswanathan, Ramesh (CNA)" <ramesh AT 
SCR.SIEMENS DOT COM> said:

> It is not recommended to have one large storage pool volume.  As each volume
> can allow only one client at any given time to write to it, you should have
> at least the same number of stg-pool volumes as the number of clients you
> will backup in parallel.  This ensures that each client gets a storage pool
> volume to dump data to on the server.


How much does this 'one volume per client' really matter?

In my exploratory preparations for implementing some disk stgpools, I decided
to test that 'one at a time' theory.  I built a stgpool with one disk volume,
and then tossed three decent-size streams at it.

As expected, they did not write to the volume at the same time; but they did
take turns very nicely.  I could watch each of the client sessions come to the
fore, more or less in turn.  It was kind of like watching a game of
hot-potato.

And this was with very fast machines, reading from fast local disk, across a
fairly-speedy network to the ADSM server.

From this, it's my impression that you can support a fairly large number of
slower, more distant clients (such as a bunch of workstations) from a single
disk volume.  No way will a dozen PCs out there be able to clog my disk.


Any counter experiences?



> Also, if you have multiple disks spread out the volumes over these disks for
> better I/O performance.

What about spreading them out by means of a logical volume assigned to (say) 8
disks?  This could let you improve the parallelism of your design without
changing the structure of stgpools, just adding disks and reorgvg.


Allen S. Rout