RE: I/O performance

From: Allen, Brandon <Brandon.Allen_at_OneNeck.com>
Date: Thu, 21 Jun 2012 17:46:56 +0000
Message-ID: <A250F0C68C23514CA9F3DF63D60EE10E163C6484_at_onews31>

Yes, considered it, but haven't actually tested it yet. I did run calibrate_io on this same machine though with the below results:

SQL> _at_calibrate_io (with num_disks=8, max_latency=100)

max_iops = 1069
latency  = 99
max_mbps = 308

I ran Windows Perfmon while calibrate_io was running to get a better idea of what it was actually doing and I saw that it ran 8k reads for the first part of the test (IOPS) and then ran 1M reads for the last part (MBPS). What was odd was that it appeared to cap itself at a latency of 3.7ms and/or queue length of 4 for the IOPS test, regardless of what I specified for num_disks or max_latency. I don't know if it was due to a bug or if it's supposed to be that way, but when I ran calibrate_io on another system, I didn't see the same behavior. So far I've only run calibrate_io on these two systems so I'm planning more testing in the future. Unfortunately there doesn't seem to be much available on the Internet about the internals of calibrate_io.

Thanks,
Brandon

-----Original Message-----

From: GG [mailto:grzegorzof_at_interia.pl]

Have You ever considered SLOB as a benchmark tool for such kind of tests ? http://kevinclosson.wordpress.com/2012/02/06/introducing-slob-the-silly-little-oracle-benchmark/

---

Privileged/Confidential Information may be contained in this message or attachments hereto. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of this company shall be understood as neither given nor endorsed by it.
--

http://www.freelists.org/webpage/oracle-l Received on Thu Jun 21 2012 - 12:46:56 CDT