If you want a third opinion, install iozone and run the benchmark on both filesystems. It does a good job of removing software and hardware caching effects from the physical read/write throughput.

-Michael

Peter Santos <psantos@(protected)-----
Hash: SHA1

Exactly,

I ran the ORION tool against the device /dev/dsk/c5t0d0s0 which is an active lun mounted as /z2. I made sure my tests
were all readonly .. no writes. This machine holds an active standby database only... and during the testing I
temporarily stopped recovery.

The filesystem is an older version of Veritas without the Quick IO option. The mount options are only delaylog, but
I'm considering remounting with mincache=direct, and convosync=direct.



- -peter



Kevin Closson wrote:
> Kevin,
> does it matter how the filesystem is mounted?
> After all I'm giving ORION a device name called
> /dev/dsk/c5t0d0s0. Does it matter what/how
> a filesystem is mounted on that device?
>
> ...I'm confused. You are running Orion on the device under the
> filesystem called /z2? I hope you don't write to it.
>
>
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGLmKqoyy5QBCjoT0RAhj/AJ9zDp9Apuyat8T7v7VNZN93cBHrpQCgmbkh
/BabI7dUaYm1ZRlkqBW0XVI=
=Sztn
-----END PGP SIGNATURE-----

--
To unsubscribe, email: suse-oracle-unsubscribe@(protected)
For additional commands, email: suse-oracle-help@(protected)
Please see http://www.suse.com/oracle/ before posting



   
---------------------------------
Ahhh...imagining that irresistible "new car" smell?
Check outnew cars at Yahoo! Autos.