Re: 2-node RAC faster than 6-node!
Date: Wed, 2 Apr 2008 20:59:17 -0400
I suspect that things might get even quicker if you leave only one node. Perhaps, it can be your final solution even? You can still run 2 node RAC and direct workload to a single node.
By the way, did you check that connections are actually routed to both nodes and they both active? If you connection load balancing is screwed you might be already running on a single machine and just didn't notice it in the moment of excitement (or disappointment, depending on your perspective).
On the performance tuning... well, I guess it's not easy to summarize it in the email format. You need someone wearing performance analyst hat to look at it. You could start from the Statspack or AWR report to have an initial idea of the workload and possible path of investigation. Alternatively, you could focus on the most important business functions but in your case they might lead you back to the cluster-wide analysis to see what's trashing your system.
On 2-Apr-08, at 5:40 PM, A Ebadi wrote:
> 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.