Home Page for the TeradataForum

Archives of the TeradataForum

Message Posted: Wed, 26 Jan 2011 @ 16:22:57 GMT

  <Prev Next>  
Next> Last>>  

Subj:   Compressing infrequent values
From:   Attila_Finta

Hi all. I have a simple question around value compression: is it better to compress all known values or only the most frequent values?

Simple example: Currency Code in a transactional table has about 50+ distinct values, but only 9 occur > 1%. All possible values for the column are known, and fewer than 255 exist, but most values occur less than 1% of the time.

     Col_Val Val_Pct
     ------- -------
     USD       59.1%
     EUR        7.6%
     CNY        6.3%
     JPY        5.0%
     BRL        3.7%
     GBP        3.3%
     CAD        2.9%
     AUD        2.6%
     INR        2.5%

I can add the other 40 values to the value list. But would that create greater operational efficiency or less? I suppose the simple answer is: a little less efficient load and a little more efficient retrieval. In general we want to optimize for retrieval. Therefore do you advise that we add all possible values? Or only the most frequent values?


Attila Finta

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