Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: cluster table/index

Re: cluster table/index

From: Erik Diehn <erikd_at_nospamm.cometsystems.com>
Date: Tue, 16 Nov 1999 16:27:45 -0500
Message-ID: <3831CC51.EB29EF02@nospamm.cometsystems.com>


This case shouldn't be the basis for creating a cluster. You may be thinking of an index-organized table, where data is stored based on a single index column (such as the PK). Of course, in that case, you have to be careful -- under Oracle 8, you can't have additional indexes on an index-organized table. I think that may have changed in 8i (anyone?).

Clusters are used primarily when two tables are often queried together and joined on a common row. I would recommend using 'order by'; if you don't want to type that all the time, create a view with the order by in it and select from that.

HTH,
Erik Diehn
Systems Engineer
erikd_at_nospamm.cometsystems.com

Larry Pettit wrote:
>
> I have a large table that contains 1.7 million rows. The primary key is a
> sequencer. New rows are always added at the end. I would like to always
> have it sorted by PK when displayed. I now that I could use 'order by', but
> would it be better to make it a cluster table and make the primary key a
> cluster index?
>
> Thanks for your help.
Received on Tue Nov 16 1999 - 15:27:45 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US