 |
 |
Archives of the TeradataForum
Message Posted: Wed, 19 Apr 2006 @ 19:16:39 GMT
Subj: | | Re: Secondary Index Space and PPI |
|
From: | | Fred W Pluebell |
In Teradata, the PI is part of the physical structure of the data portion of the table; there is no separate "primary index". Unless the
partitioning expression columns are all included within the PI, verifying PI uniqueness would require probing all partitions.
You can define a USI with the same set of columns as the NUPI (allowing efficient uniqueness checking) if your application requires that.
| |