Archives of the TeradataForum

Message Posted: Mon, 17 Mar 2008 @ 22:51:49 GMT

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

<-- Anonymously Posted: Mon, 17 Mar 2008 21:34 -->

Hi,

I need to create some evenly distributed dummy data tables which will be used to reserve diskspace for capacity on demand. My method for creating these tables was to use a unique primary index populated with integer values in a contiguous number sequence e.g.

     CREATE SET TABLE CAPACITY_ON_DEMAND.COD_1 ,NO FALLBACK ,
     NO BEFORE JOURNAL,
     NO AFTER JOURNAL,
     CHECKSUM = DEFAULT
     (
     seq_no FLOAT,
     Col1 CHAR(32000) CHARACTER SET LATIN NOT CASESPECIFIC,
     Col2 CHAR(32000) CHARACTER SET LATIN NOT CASESPECIFIC)
     UNIQUE PRIMARY INDEX ( seq_no );

After populating this table with data I ran the following queries which all returned the same value :-

     select max(seq_no) from CAPACITY_ON_DEMAND.COD_1
     select count(*) from CAPACITY_ON_DEMAND.COD_1
     select count(distinct(seq_no )) from CAPACITY_ON_DEMAND.COD_1

However the data in this table has a skew factor of 2. Can anyone tell me why the skew factor is not 0 ? Does anyone have an alternative method which will create a table which has a data skew factor of 0 ?

Many thanks

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