thanks mladen. we removed one of the disks from asm and are running iozone
on it to see if the numbers sync up with what we see through oracle.
On Mon, May 7, 2018 at 11:57 AM Mladen Gogala <gogala.mladen@xxxxxxxxx>
wrote:
With all due respect, expensive clusters like that are usually
engineered for performance when the hardware is chosen. You should have
chosen 32Gbit FC adapters and either an all flash array like Xtrem_IO or
Pure or a disk array with lots of cache. You should have calibrated your
IO system for the needed number of IOPS. There are lots of cheap tools
to do that, bonnie++ and iozone3 among the most popular ones. SLOB is a
great tool, but for SLOB you need to have a database already installed.
Once SLOB discovers that your IO is slow, there is very little you can
do in terms of the underlying hardware.
On 05/07/2018 09:59 AM, Chris Stephens wrote:
We have a new 5 node 12.2 RAC system that is not performing the way we
want.
The glaring issue is that "db file sequential read"'s are taking
~10ms. before i lob this over to the storage administrators, are there
any possible areas in the clusterware/database configuration that I
should investigate first? i have root access to all of the nodes. is
there any information i can collect that would expedite the process of
figuring out why we have such slow I/O times?
slow i/o was discovered by running slob. if you don't know about that
tool, you should. we all owe kevin a debt of gratitude. ;)
if nothing else, i hope to learn a little more about storage than i
currently know (which isn't much).
thanks for any help.
chris
--
Mladen Gogala
Database Consultant
Tel: (347) 321-1217
--
//www.freelists.org/webpage/oracle-l
