From: | "Jim C(dot) Nasby" <jim(at)nasby(dot)net> |
---|---|
To: | "scott(dot)marlowe" <scott(dot)marlowe(at)ihs(dot)com> |
Cc: | Matthew Nuzum <cobalt(at)bearfruit(dot)org>, Pgsql-Performance <pgsql-performance(at)postgresql(dot)org> |
Subject: | Re: choosing the right platform |
Date: | 2003-04-22 08:30:58 |
Message-ID: | 20030422033058.B37876@flake.decibel.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On Thu, Apr 10, 2003 at 10:42:35AM -0600, scott.marlowe wrote:
> The difference between SIZE and SHARE is the delta, which is only
> something like 3 or 4 megs for the initial select * from logs, but the
> second one is only 1 meg. On average, the actual increase in memory usage
> for postgresql isn't that great, usually about 1 meg.
>
> Running out of memory isn't really a problem with connections<=200 and 1
> gig of ram, as long as sort_mem isn't too high. I/O contention is the
> killer at that point, as is CPU load.
Except you should consider what you could be doing with that 200M, ie:
caching data. Even something as small as 1M per connection starts to add
up.
--
Jim C. Nasby (aka Decibel!) jim(at)nasby(dot)net
Member: Triangle Fraternity, Sports Car Club of America
Give your computer some brain candy! www.distributed.net Team #1828
Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"
From | Date | Subject | |
---|---|---|---|
Next Message | Christopher Kings-Lynne | 2003-04-22 08:33:27 | Re: [SQL] Yet Another (Simple) Case of Index not used |
Previous Message | Jim C. Nasby | 2003-04-22 08:23:39 | Re: [SQL] Yet Another (Simple) Case of Index not used |