Fwd: Interesting performance behaviour - Mailing list pgsql-performance

From Joey Smith
Subject Fwd: Interesting performance behaviour
Date
Msg-id e41f745b04091015023ef1fabc@mail.gmail.com
Whole thread Raw
In response to Interesting performance behaviour  (Joey Smith <joeysmith@gmail.com>)
List pgsql-performance
Accidentally sent directly to Josh.


---------- Forwarded message ----------
From: Joey Smith <joeysmith@gmail.com>
Date: Fri, 10 Sep 2004 15:57:49 -0600
Subject: Re: [PERFORM] Interesting performance behaviour
To: josh@agliodbs.com

> > I see a similar speedup (and change in query plan) using "LIMIT 1
> > OFFSET <anything besides 0>".
>
> So what's your problem?

The problem is that "LIMIT 1 OFFSET 0" has such poor performance. I'm
not so much worried about the query time (it's still low enough to be
acceptable), but the fact that it behaves oddly raised the question of
whether this was correct behaviour or not. I'll try it with a saner
value for effective_cache_size.

pgsql-performance by date:

Previous
From: Joey Smith
Date:
Subject: Re: Interesting performance behaviour
Next
From: Josh Berkus
Date:
Subject: Re: effective_cache_size in 7.3.4?