Question about use of DISTINCT
From: Sandra Becker <sbecker6925_at_gmail.com>
Date: Mon, 17 Oct 2011 12:49:01 -0600
Message-ID: <CAJzM94Aj-tkGd-sF5X8dz5YBkmXyBntWHYmjUTobdxUSDdCL5g_at_mail.gmail.com>
IBM zSeries
SUSE 10
Oracle EE 10.2
A developer recently put some code into production that I did not get to see until 3 days later when it was failing part way through and causing problems with the other part. One of the things I noticed was the use of DISTINCT on columns that are already unique. Because of the predicates and the way the primary keys were defined, they will never use the primary key index or any other unique index on the tables.
Date: Mon, 17 Oct 2011 12:49:01 -0600
Message-ID: <CAJzM94Aj-tkGd-sF5X8dz5YBkmXyBntWHYmjUTobdxUSDdCL5g_at_mail.gmail.com>
IBM zSeries
SUSE 10
Oracle EE 10.2
A developer recently put some code into production that I did not get to see until 3 days later when it was failing part way through and causing problems with the other part. One of the things I noticed was the use of DISTINCT on columns that are already unique. Because of the predicates and the way the primary keys were defined, they will never use the primary key index or any other unique index on the tables.
Question: Is using the DISTINCT on a column that is already unique causing unnecessary overhead or is it just superfluous and the database is smart enough to ignore it? I've run a few explain plans and the use of the DISTINCT does not change the explain plan.
--
Sandy
Transzap, Inc.
--
http://www.freelists.org/webpage/oracle-l
Received on Mon Oct 17 2011 - 13:49:01 CDT