partitioned table advice

From: Ed Stevens <Ed_Stevens_at_nospam.noway.nohow>
Date: Thu, 13 Dec 2001 23:59:27 GMT
Message-ID: <3c193f41.87010724@ausnews.austin.ibm.com>

Platform; Oracle 8.0.5 SE, NT4

We've never used a partitioned table before, but I'm looking at a candidate. A SELECT DISTINCT on the primary key shows seven different values. The distribution of rows across these values is as follows:

value 1 - 17,814 rows
value 2 -      322 rows
value 3 - 36,976 rows
value  4 - 29,580 rows

value 5 - 228 rows
value 6 - 1,565 rows
value 7 - 371 rows

The app is heavy in queries with a "WHERE =" on this column.

Would the skewing of the column values argue for or against partitioning on this column? Or is the answer still "it depends -- we need more info about which distinct values are most often queried."

--
Ed Stevens
(Opinions expressed do not necessarily represent those of my employer.)

Received on Thu Dec 13 2001 - 17:59:27 CST