>Then when creating the Database volumes create multiple volumes per
physical
>drive. If you've been reading the list a lot of people have pointed out
that
>a separate thread is allocated for each database volume, therefore the
>larger the number of database volumes the greater the throughput. However
>there must be some sort of limit to this...

I would think that the # of volumes you want to create would be dependant on
how much is going on.  If you've got 30 nodes backing up at the same time,
then it would be helpful to have a thread per node available in your disk.
Of course, that should probably be tempered a bit by head contention... for
example if you have those 30 volumes all on 2 physical disks, my guess would
be that you're going to experience a performance drop due to the read write
heads flipping around between volumes.
I don't know at what point the trade off comes with the volumes/disk ratio.
Does anyone have any stats they've found on that one?

Of course, if you're only backing up 2 nodes at one time, I would think that
1 volume on each of those physical disks should work just fine and
dandy.....

Paul