RE: 2-node RAC faster than 6-node!

From: Bobak, Mark <>
Date: Wed, 2 Apr 2008 18:21:32 -0400
Message-ID: <>

Hi Abdul,

It's tough to say, without really knowing what's going on. I will say this, though: Rule #1 about adding hardware when faced w/ a performance problem is that if you add capacity to a non-bottleneck resource, there's a good chance you'll make the performance worse!

So, consider what your bottleneck really was, and whether you didn't make the bottleneck worse. For example, what are you using for your interconnect? What if that was already saturated, and adding extra nodes just put more load on the interconnect, further saturating it?

Total speculation, without knowing more about your system, I can't say....but it's just one hypothesis.



Mark J. Bobak
Senior Database Administrator, System & Product Technologies ProQuest
789 E. Eisenhower, Parkway, P.O. Box 1346 Ann Arbor MI 48106-1346
+1.734.997.4059 or +1.800.521.0600 x 4059<><><>

ProQuest...Start here.

From: [] On Behalf Of A Ebadi Sent: Wednesday, April 02, 2008 5:41 PM
Subject: 2-node RAC faster than 6-node!

We have an existing 4-node RAC cluster (v440's on Solaris & Oracle 10.2 w/ ASM) that was big time CPU-bound and application performance was suffering. We added 2 new nodes(M5000's) to bring the total to 6 and the app performance actually went down significantly! After messing around with it we found out that by taking the original 4 nodes out of the picture & only running on the 2 new M5000's nodes things improved significantly - the app jobs started running in some cases 20X faster than ever! We tested adding just a single node (v440) back in the mix and performance went way down immediately, so we decided to keep things running on just the 2 M5000's for now.

Any suggestions on identifying root cause as to why the adding of any nodes back to the picture slows things way down? We have an SR opened with Oracle Support but nothing firm so far.


You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access<*http:/>, No Cost.

-- Received on Wed Apr 02 2008 - 17:21:32 CDT

Original text of this message