From: Neil Tompkins
Date: October 11 2011 1:00pm
Subject: Re: Index question
List-Archive: http://lists.mysql.com/mysql/226010
Message-Id: <345A4C5C-1107-4F31-96A8-08A7935C445C@googlemail.com>
MIME-Version: 1.0 (iPhone Mail 8L1)
Content-Type: text/plain;
	charset=us-ascii
Content-Transfer-Encoding: quoted-printable

In this instance would you create four indexes key(a) key(b) key(a,b) key (b=
,a) ? Or is the decision based on the query response time ?

On 11 Oct 2011, at 13:40, Rik Wasmus <rik@stripped> wrote:

>> Next question. If you have the two separate indexes and then do two
>> queries, one for a and one for b. If you then get a list of unique id's
>> of both, would it be faster to create an intersection yourself rather
>> than have the server do the legwork?
>=20
> If you only have 2 unrelated indexes on a & b, it depends on the data, the=
=20
> distribution of values, etc. No single answer here, test with your data an=
d=20
> you'll have the results.
>=20
> If you need it often, I'd go for the combined index & let MySQL do the wor=
k,=20
> which is probably fastest.=20
> --=20
> Rik Wasmus
>=20
> --=20
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe:    http://lists.mysql.com/mysql?unsub=3Dneil.tompkins@goog=
lemail.com
>=20