 |
 |
Archives of the TeradataForum
Message Posted: Mon, 10 Jun 2002 @ 20:47:43 GMT
Subj: | | Re: Full-Table Scans and Secondary Indexes |
|
From: | | Raghav Vaidya |
Clay,
Teradata stores base tables rows in different blocks than the sub-table rows. In a read, only those blocks are read that have the base
table rows required to satifsy the query. In case of a FTS on a table with NUSI, only those blocks with base table rows are read while on
tables with a USI, blocks with rows based on the USI are read.
I believe the time taken should be same.
| |