To: Kevin Brown <kevin@sysexperts.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: [SQL] Yet Another (Simple) Case of Index not used 
In-reply-to: <20030419130146.GI1847@filer> 
References: <EKEBJNAJDPKJBDFGJNIJAENMDBAA.denis@next2me.com>
	<200304090918.45275.josh@agliodbs.com>
	<20030419130146.GI1847@filer>
Comments: In-reply-to Kevin Brown <kevin@sysexperts.com>
	message dated "Sat, 19 Apr 2003 06:01:46 -0700"
Date: Sat, 19 Apr 2003 11:58:48 -0400
Message-ID: <21955.1050767928@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Kevin Brown <kevin@sysexperts.com> writes:
> Josh Berkus wrote:
>> They did incorporate a lot of MVCC logic into InnoDB tables, yes.
>> Which means that if SELECT count(*) on an InnoDB table is just as
>> fast as a MyISAM table, then it is not accurate.

> This is not necessarily true.  The trigger-based approach to tracking
> the current number of rows in a table might well be implemented
> internally, and that may actually be much faster than doing it using
> triggers

You missed the point of Josh's comment: in an MVCC system, the correct
COUNT() varies depending on which transaction is asking.  Therefore it
is not possible for a centrally maintained row counter to give accurate
results to everybody, no matter how cheap it is to maintain.

(The cheapness can be disputed as well, since it creates a single point
of contention for all inserts and deletes on the table.  But that's a
different topic.)

			regards, tom lane