I do have a dirty little secret, one which I wasn't completely aware
of until a little while ago. Apparently, someone decided to install
Oracle on the server, and use the SAN as the primary tablespace, so
that might have something to do with the poor performance of the SAN.
At least, I'm hoping that's what is involved. I'll be able to tell
for sure when I can rerun the benchmarks on Monday without Oracle.
Thank you all for all your help so far. I've learned a ton about
Postgres (and life!) from both this thread and this list in general,
and the "support" has been nothing less than spectacular.
I'm hoping that the corporate Oracle machine won't shut down my pg
projects. On total side note, if anyone knows how to best limit
Oracle's impact on a system (i.e. memory usage, etc), I'd be
I hate shared DB servers.
In response to
pgsql-performance by date
|Next:||From: Merlin Moncure||Date: 2006-10-27 19:01:51|
|Subject: Re: Best COPY Performance|
|Previous:||From: Tom Lane||Date: 2006-10-27 18:07:43|
|Subject: Re: VACUUMs take twice as long across all nodes |