Re: RAC or Large SMP...?
Date: Tue, 07 Oct 2008 06:49:47 -0700
> I support an OLTP application which handles 2 million transactions per
> day and is running on 10gR2 EE on RHEL 4 x86_64.
> I am investigating scaling options for the application and I'm trying
> to decide between 2 large SMP servers or a multi-node RAC
> As far as I can tell, the highest number of cores available in an
> x64_64 server is 24. This would only allow us to handle 6 times the
> current workload, and realistically we need to be able to support up
> to 20 times the load.
> Has anyone had any experience of comparing the 2 approaches with
> respect to cost, manageability, performance, etc. Can you offer any
> advise and/or pointers to resources to help out with this
> Specifically I'm interested in:
> What is the most powerful x86_64 machine available..?
> Can Oracle scale well on NUMA based architectures (as some of the high
> end x64_64 based servers seem to be)..?
> Is the cost difference between '2 x large SMP' and 'multi-node RAC'
> large enough to justify the extra complexity of administering a
> cluster environment...?
> Any assistance on this would be greatly appreciated.
The decision should rest, at least in part, in an analysis of the code being run and what it is doing.
RAC is a good solution if you can isolate different types of workload using services to different nodes. Also important is code and usage that minimizes block, lock, and sequence sharing and maximizes what you will find in the Oracle literature described as node affinity.
For example: To optimize AQ on a RAC cluster put a different queue on each node.
Earlier this year I ran an AQ class for Dell in which I demoed AQ on RAC live. The performance difference was with node affinity, versus without node affinity, could have been seen by someone that was blind. <g>
-- Daniel A. Morgan Oracle Ace Director & Instructor University of Washington damorgan_at_x.washington.edu (replace x with u to respond) Puget Sound Oracle Users Group www.psoug.orgReceived on Tue Oct 07 2008 - 08:49:47 CDT