Re: using limit

Is a table with 10 million of rows with a primary key but conditions used in
select are not part of PK.

My table is like this:

Column 1 - PK

Column 2 - Indexed

Column 3 - Indexed

Column 2 and Column 3 are Indexed in the same index but they are not mark as
unique or PK

The Query is like this:

Select column1 from myTable where Column2 between X and Y

I am expecting just one record

De: Michael Holt [mailto:michael(at)aers(dot)ca]
Enviado el: jueves, 16 de diciembre de 2010 04:23 p.m.
Para: Anibal David Acosta; pgsql-admin(at)postgresql(dot)org
Asunto: RE: [ADMIN] using limit

This really depends on the type of query you're talking about. If there's
only one row in the table you're querying then no, I don't think it'll
change anything. If you're querying a single row using a primary key it
shouldn't change anything. If you're doing an aggregate query, say a sum of
a bunch of rows, it also won't improve performance.

If you're doing a query on a table with multiple rows and not filtering by a
primary key or other unique index then yes, it will improve the query.

From: pgsql-admin-owner(at)postgresql(dot)org
[mailto:pgsql-admin-owner(at)postgresql(dot)org] On Behalf Of Anibal David Acosta
Sent: Thursday, December 16, 2010 11:19 AM
To: pgsql-admin(at)postgresql(dot)org
Subject: [ADMIN] using limit

Hi

I really like to do efficient SQL queries so, my question is if I am
expecting no more than one row from a select, using the LIMIT 1 could
improve the performance?

If I use my logic, the LIMIT 1 instruction tell to postgres that stop
searching when found 1 record, but maybe it is unnecessary

Thanks

Anibal