Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Sat, 10 Jan 2015 @ 16:51:15 GMT


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


Subj:   Restructure large tracking table
 
From:   Anomy Anom

<-- Anonymously Posted: Saturday, January 10, 2015 09:20 -->

Happy New Year.

I would appreciate any advice on how to improve performance against an ever growing table.

It tracks changing measures against a customer's account on daily basis. (in excess of 1 billion rows) and has a single NUPI on the Account_No:

     Account_Number (NUPI)
     Effective_From_Date
     Effective_To_Date
     Measure_Amount
     etc

Typical queries are 'what is the Amount at the end of the month?' These result in full table scans that are taking longer and longer on what is a highly used table.

Are there any effective ways to index or structure such a table? I have considered PPI but don't see how that could be implemented. Adding PPI to the To_Date would result in movement across partitions each time a new row was added.

I have read about the new 'temporal' data types in V13 however I am reluctant to have a mixture of table structures and inherent SQL usage in my database or is it worth considering for a single table?

Any advice appreciated.



     
  <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: 27 Dec 2016