Re: mis-estimation on data-warehouse aggregate creation - Mailing list pgsql-performance

From Kris Jurka
Subject Re: mis-estimation on data-warehouse aggregate creation
Date
Msg-id Pine.BSO.4.56.0411170315050.13112@leary.csoft.net
Whole thread Raw
In response to Re: mis-estimation on data-warehouse aggregate creation  (Simon Riggs <simon@2ndquadrant.com>)
List pgsql-performance

On Tue, 16 Nov 2004, Simon Riggs wrote:

> The join condition has so many ANDed predicates that we assume that this
> will reduce the selectivity considerably. It does not, and so you pay
> the cost dearly later on.
>

Yes, that makes a lot of sense.  Without some incredibly good cross-column
statistics there is no way it could expect all of the rows to match.
Thanks for the analysis.

Kris Jurka

pgsql-performance by date:

Previous
From: Greg Stark
Date:
Subject: Re: memcached and PostgreSQL
Next
From: "David Parker"
Date:
Subject: Re: query plan question