Home Page for the TeradataForum

Archives of the TeradataForum

Message Posted: Mon, 16 Aug 2004 @ 20:43:54 GMT

  <Prev Next>   <<First <Prev Next> Last>>  

From:   Dieter Noeth

Christie, Jon wrote:

  Even if the target table is initially empty, duplicate row checking will still have to be performed. Why? Because the UNION only guarantees that there will be no duplicate rows for ALL COLUMNS. This does not guaranteed that some subset of columns (like the PI of the target table) will be unique.  

Hi Jon, you probably just confused "Duplicate Rows" with "NUPI duplicates" ;-)

The essence of your first two sentences:

"duplicate row checking have to be performed because UNION guarantees that there will be no duplicate rows"

  By the way, duplicate checking is only expensive where there are lots of duplicates.  

Duplicate row checks are expensive if there are lots of NUPI duplicates.

  If it turns out that the spool really is unique on the PI of the target table, the overhead for duplicate checking will very, very small.  



  <Prev Next>   <<First <Prev Next> Last>>  
  Top Home Privacy Feedback  
Copyright for the TeradataForum (TDATA-L), Manta BlueSky    
Copyright 2016 - All Rights Reserved    
Last Modified: 15 Jun 2023