Archives of the TeradataForum

Message Posted: Fri, 12 Jan 2007 @ 11:38:07 GMT

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

Hi

I'm having a table like below ,

     CREATE SET TABLE 
          (
           TableCode CHAR(5),
           TableData VARCHAR(995),
           loaddate DATE)
     PRIMARY INDEX ( TableCode );

The Table is set with out a unique primary index , and contains the combination of the two columns as a unique value . there is a skew
factor of 96% on the table . the table is being used in joins.
If the table is having the Second column as a primary Index , there will not be any Skew in the data .
Now the Questions are
is there any Performance issue due to the skew being 96 considering the table is small (just 21,000 Rows)
if the Second column , which cant be used in the queries since it's a description of the Code , be used as a primary index , and
there is no skew in data , will it improve the performance ???
3. many of my Production tables have a skew factor of 30 -45 % . Will there be performance Bottelneck due to this ? ? ?

Thanks in Advance
Adurthy Sri Krishna


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