RE: Oracle Performance on VMWare

From: Allen, Brandon <Brandon.Allen_at_OneNeck.com>
Date: Fri, 2 Jan 2009 12:41:20 -0700
Message-ID: <64BAF54438380142A0BF94A23224A31E10A32F1852_at_ONEWS06.oneneck.corp>



Unfortunately I didn't have time to document it very clearly for others, but I did a lot of testing with ORION and dd to run multiple IO channels concurrently and did some testing with Swingbench and some basic custom Oracle I/O scripts (just a few basic concurrent index scans and full table scans on large test tables). We're running Dell 2950s with dual quad-core Intel Xeon X5450 3.0GHz procs, 32GB RAM, fiber attached to CX3-40 disk array with RAID 5 groups striped across 5 disks and we were maxing out at about 900 IOPS with both bare metal and on top of VMWare. One key thing to note is that we were running just a single VM per physical box so maybe part of the problem is that VMWare suffers when it has to manage the IO between multiple VMs?

Some might wonder - why run just a single VM per box? This was done for high availability, not server consolidation. We have a cluster of several VMWare servers running Oracle DB, App Server, MS Active Directory, Citrix, etc. - and we can easily move VMs from box to box whenever we want. Some of the boxes do run multiple VMs for other applications, but for the Oracle DBs, it's just 1-to-1.

Regards,
Brandon

From: Crisler, Jon [mailto:Jon.Crisler_at_usi.com]

I would be interested in reading about how you set up your benchmarks.



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 Fri Jan 02 2009 - 13:41:20 CST

Original text of this message