Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Thu, 25 Jun 2009 @ 10:35:56 GMT


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


Subj:   Re: Partitioning Start_Date End_Date pair for Dimensions
 
From:   David Clough

I think David Hurley has just about given the definitive answer to your question but I can tell you what we do.

I've taken it upon myself to construct a process that, after the SCD (Slowly Changing Dimension) has been loaded, we have a final step that takes the 'current' entries from the SCD and populate another table. So, effectively, we have two physical tables : a History table (the SCD) and a 'Current' table.

The code to do this has been auto-generated (using Erwin script) so doesn't cause any anxiety for the developers.

We've done this because, unlike in your case, we normally only require the current entry from the Dimension, not the historical view. I wanted to ensure the optimizer had the best chance of determining the number of rows from a join to the Dimension, without having to work out 'Where Fact_Depot = Dim_Depot and Fact_Date between Dim_Date_Fr and Dim_Date_To' type stuff. (I've yet to be thanked for doing this, but I'm hopeful !)

'Persuading' the users that the 'current' view is the best view, makes life a lot easier !


Regards

David Clough
Database Developer
Database Design Group



     
  <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