Archives of the TeradataForum
Message Posted: Fri, 25 Apr 2003 @ 14:10:32 GMT
<-- Anonymously Posted: Friday, April 25, 2003 10:02 -->
Note - to make the samples unique (no duplicate rows) the read starts with the first row and reads sequentially (so to speak). Using a probability algorithm that uses a random number between 0 and 1 (obtained by UNIX "drand" or NT "rand") we place the row in some sample. The rows from all the amps are returned the user.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|