Archives of the TeradataForum
Message Posted: Mon, 28 Mar 2016 @ 09:53:41 GMT
<-- Anonymously Posted: Monday, March 28, 2016 04:44 -->
What is/are the reason(s) why a PK will be implemented as USI instead of UPI? This question was formulated based on the following statement:
"Primary keys are implemented as unique secondary indexes for non-temporal tables..."
Reference: "Database Design" (Release 14.0); page 296
I think that PK as UPI is better since it will uniformly distribute the rows across all AMPs since PK is always unique by definition.
So why use USI for a PK?
Thank you in advance for any responses.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 23 Jun 2019|