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

Re: Performance of count(*)

From: Mario Weilguni <mweilguni(at)sime(dot)com>
To: pgsql-performance(at)postgresql(dot)org
Cc: ismo(dot)tuononen(at)solenovo(dot)fi, Albert Cervera Areny <albert(at)sedifa(dot)com>
Subject: Re: Performance of count(*)
Date: 2007-03-22 13:26:51
Message-ID: 200703221426.51948.mweilguni@sime.com (view raw or flat)
Thread:
Lists: pgsql-performance
Am Donnerstag, 22. März 2007 12:30 schrieb ismo(dot)tuononen(at)solenovo(dot)fi:
> approximated count?????
>
> why? who would need it? where you can use it?
>
> calculating costs and desiding how to execute query needs
> approximated count, but it's totally worthless information for any user
> IMO.

No, it is not useless. Try: 
http://www.google.com/search?hl=de&q=test&btnG=Google-Suche&meta=

Do you really think google counted each of those individual 895 million 
results? It doesn't. In fact, the estimate of google can be off by an order 
of magnitude, and still nobody complains...


In response to

pgsql-performance by date

Next:From: DimitriDate: 2007-03-22 13:35:53
Subject: Parallel Vacuum
Previous:From: Bill MoranDate: 2007-03-22 13:17:00
Subject: Re: Potential memory usage issue

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