Skip site navigation (1) Skip section navigation (2)

Re: More thoughts about planner's cost estimates

From: Josh Berkus <josh(at)agliodbs(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Cc: Greg Stark <gsstark(at)mit(dot)edu>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Subject: Re: More thoughts about planner's cost estimates
Date: 2006-06-01 21:45:40
Message-ID: 200606011445.41158.josh@agliodbs.com (view raw or flat)
Thread:
Lists: pgsql-hackers
Greg,

> > 1) You have n^2 possible two-column combinations. That's a lot of
> > processing and storage.
>
> Yes, that's the hard problem to solve.  Actually, btw, it's n!, not n^2.

Ooops, bad math.  Andrew pointed out it's actually n*(n-1)/2, not n!.

Also, we could omit columns unlikely to correlate, such as large text 
columns, bytea and numerics with high precisions.  Also, we probably don't 
need to correlate UNIQUE columns inside ... I think.

-- 
--Josh

Josh Berkus
PostgreSQL @ Sun
San Francisco

In response to

pgsql-hackers by date

Next:From: Tom LaneDate: 2006-06-01 21:53:14
Subject: Re: pgsql: Fix calculation of plan node extParams to account for the
Previous:From: Kenneth MarshallDate: 2006-06-01 21:33:09
Subject: Re: More thoughts about planner's cost estimates

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group