On Tuesday, Apr 15, 2003, at 00:00 Australia/Sydney, Matt Simpson wrote:

At 8:16 AM -0005 4/14/03, Richard L. Rhodes wrote:

Our current TSM server uses Shark storage.  We run a 80gb TSM db on 2
shark raidsets (8 packs) of 18gb drives.   All 20 db volumes are in
this same filesystem along with the
log(5gb), spread across all 18 spindles.

A lot of this stuff is still pretty hazy to me, but does that mean
you're not using raw volumes for the DB?  IBM recommended that we use
raw volumes for performance reasons.  We're also running on a Shark.
We have a 60gb database spread across 9 dbvolumes, all raw.
Considering how much Shark I/O is virtual, would we be better off
defining file systems on those partitions and defining multiple DB
vols per partition?

I'd always recommend raw - stops the kernel wasting CPU cycles managing
the buffer cache, and bypasses all the normal filesystem overhead.

Just make sure you have enough DB volumes to keep the disks busy. See
my benchmarks earlier in this thread.

Cheers,
--
Paul Ripke
Unix/OpenVMS/TSM/DBA
101 reasons why you can't find your Sysadmin:
68: It's 9AM. He/She is not working that late.
-- Koos van den Hout