Well that is the rub. I am currently using Postgres everyday. It runs
like a champ and even on an old PIII 600 MHZ machine with a single old
and slow IDE drive and 256 megs of ram it is fast enough for what we do.
This is more for me to do some testing with.
I think it would useful as a tunning tool if nothing else.
What I would like to try just for my own amusment is to build a small
test box. It will not be a server class machine. I am thinking of using
an AMD X2 and to start a SATA hard drive.
Then I would like to test different file systems, then different
operating systems, different amounts of ram, 32 vs 64 bit, and software
raids.
I would use the same machine for all the tests so it would have a good
base line.
I doubt that would ever publish my results. The flame war that would
happen would take all the fun out of it for me. I am sure that someone
would say that since I wasn't using a server machine that my results
where invalid, others would say that I made errors in tuning for the
different operating systems or that X would show benefits if I was using
a real server machine. And all of them may be right.
But sometimes you just want to play.
Do you want to use PostgreSQL for any particular task? If so, the best
benchmark is probably one that simulates your specific workload.
Comparing generic benchmark results like pgbench etc may not usefully
reflect performance in real-world use with your load and your data.
--
Craig Ringer
