Date: Fri, 27 Jul 2007 16:03:11 -0400
From: Lew <lew@lewscanon.nospam>
User-Agent: Thunderbird 2.0.0.5 (X11/20070719)
MIME-Version: 1.0
Subject: Re: multicolumn index column order
References: <200707240710.l6O7AUu24090@pillette.com>
	<3ce9822f0707240040o2f358155p7ff7537bbe986209@mail.gmail.com>
	<ZcadnVxk8I9ICDvbnZ2dnUVZ_gidnZ2d@comcast.com>
	<1185352440.504985.282370@19g2000hsx.googlegroups.com>
In-Reply-To: <1185352440.504985.282370@19g2000hsx.googlegroups.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Message-ID: <tqCdnQLIFqLi0jfbnZ2dnUVZ_u3inZ2d@comcast.com>
Lines: 24
To: pgsql-performance@postgresql.org

valgog wrote:
> On Jul 25, 2:14 am, Lew <l...@lewscanon.nospam> wrote:
>> How about two indexes, one on each column?  Then the indexes will cooperate
>> when combined in a WHERE clause.
>> <http://www.postgresql.org/docs/8.2/interactive/indexes-bitmap-scans.html>
>>
>> I don't believe the index makes a semantic difference with regard to ascending
>> or descending.  An index is used to locate records in the selection phase of a
>> query or modification command.
>>
>> --
>> Lew
> 
> Ordered indexes (b-tree in this case) are also used to get the needed
> record order and it is absolutely not necessary to have a WHARE clause
> in your select statement to use them when you are using ORDER BY.

But does that affect anything when you "ORDER BY foo ASC" vs. when you "ORDER 
BY foo DESC"?

For use by ORDER BY, separate column indexes are an even better idea.

-- 
Lew