Stan Hoeppner stan@hardwarefreak.com wrote:
The more I think about your planned architecture the more it reminds me of a "shared nothing" database cluster--even a relatively small one can outrun a well tuned mainframe, especially doing decision support/data mining workloads (TPC-H).
As long as you're prepared for the extra administration, which you obviously are, this setup will yield better performance than the NFS setup I recommended. Performance may not be quite as good as 4 physical hosts with local storage, but you haven't mentioned the details of your SAN storage nor the current load on it, so obviously I can't say with any certainty. If the controller currently has plenty of spare IOPS then the performance difference would be minimal.
This is the beauty of the HP P4500: every node is a controller, load is automagically balanced between all nodes of a storage cluster. The more nodes (up to ten) you add, the more performance you get.
So far, I have not been able to push our current SAN to its limits, even with totally artificial benchmarks, so I am quite confident in its performance for the given task.
But if everything fails and the performance is not good, I can still go ahead and buy dedicated hardware for the mailsystem.
The only thing left is the NFS problem with caching Timo mentioned, but since the accesses to a central public shared folder will be only a minor portion of a clients access, I am hoping the impact will be minimal. Only testing will tell.
Grüße, Sven.
-- Sigmentation fault. Core dumped.