Archives of the TeradataForum
Message Posted: Thu, 22 Jan 2004 @ 11:48:25 GMT
Does anyone know of a good white paper to discuss the usage of indexes and statistics. I would like something that would tell me the recommendations of when to build another index on a table as opposed to just collecting statistics on a group of columns. I have an environment that people do a lot of ad hoc queries and I want to optimize my tables the best way possible.
Currently I have used the following strategy.
Obviously the primary index. I decide this by mostly dispersion of my data across amps.
I use secondary indexes based on columns I know are join columns to other tables.
I collect stats on each column in the table individually.
I collect stats on each index in the table.
From what I learned from my last post, a view will NOT use indexes or statistics IF it is set up any differently than just CREATE VIEW AS SELECT * FROM MY_TABLE
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|