Re: Seqscan rather than Index - Mailing list pgsql-performance

From Steinar H. Gunderson
Subject Re: Seqscan rather than Index
Date
Msg-id 20041218134540.GA25107@uio.no
Whole thread Raw
In response to Re: Seqscan rather than Index  (Bruno Wolff III <bruno@wolff.to>)
List pgsql-performance
On Fri, Dec 17, 2004 at 10:39:18PM -0600, Bruno Wolff III wrote:
> It doesn't seem totally out of wack. You will be limited by the memory
> bandwidth and it looks like you get something on the order of a few
> hundred references to memory per row. That may be a little high, but
> it doesn't seem ridiculously high.

I just tested 8.0.0rc1 -- I got a _50%_ speedup on this operation...

/* Steinar */
--
Homepage: http://www.sesse.net/

pgsql-performance by date:

Previous
From: Stan Y
Date:
Subject: Monitor CPU time per transaction?
Next
From: Theo Galanakis
Date:
Subject: PG Logging is Slow