Re: 8.1 count(*) distinct: IndexScan/SeqScan - Mailing list pgsql-performance

From Tom Lane
Subject Re: 8.1 count(*) distinct: IndexScan/SeqScan
Date
Msg-id 7006.1132802096@sss.pgh.pa.us
Whole thread Raw
In response to 8.1 count(*) distinct: IndexScan/SeqScan  (Pailloncy Jean-Gerard <jg@rilk.com>)
Responses Re: 8.1 count(*) distinct: IndexScan/SeqScan
List pgsql-performance
Pailloncy Jean-Gerard <jg@rilk.com> writes:
> Why the stupid indexscan plan on the whole table ?

Pray tell, what are you using for the planner cost parameters?
The only way I can come close to duplicating your numbers is
by setting random_page_cost to somewhere around 0.01 ...

            regards, tom lane

pgsql-performance by date:

Previous
From: "Luke Lonergan"
Date:
Subject: Re: Hardware/OS recommendations for large databases (
Next
From: Mark Kirkwood
Date:
Subject: Re: Hardware/OS recommendations for large databases (