So Velocibunny now has a shiny new set of 5502 Nehalem CPUs in it, 12 GB ram, and 24 SSDs.
For laughs, I ran BenchmarkSQL on it. Ok, not for laughs, but the folks who were originally all hot an bothered to run on it sort of disappeared, so I had to come up with some benchmark tests. Oddly enough, BenchmarkSQL was written by one of them. Go figure.
I am attempting to understand, in the simplest possible classification scenario, what a good score is and what a bad score is. BenchmarkSQL returns something that is supposed to be similar to a tpc-c tpmC result. I don’t know how comparible it is, but thats what it is supposed to do.
So I ran it in RAID10 and RAID6 mode.
And I have numbers.
But I can’t seem to find things to compare to. Which suggests that I don’t quite understand the test. And this annoys me. Because it also means that I don’t know how to tune the system or the Postgres database that is now running on the SSDs.
Additionally, it appears that this is a java application. I generally have lots of reservations about using java where performance matters … granted the DB is likely the bottleneck here, but still …
So I have 4 cores of Nehalem, 12 GB ram, and a bloody fast set of SSD hooked up. Assume I have 4 more days of testing. What should I test?
Please let me know what you think. File system benches … database benches … all open for consideration.