Archives of the TeradataForum

Message Posted: Wed, 16 Jul 2003 @ 12:09:19 GMT

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

some options may be

1. build a summary table that is at the level of most of the group by's

2. make sure users are stongly selective in there "where" clause - hopefully on an indexed column. (make sure that you take full advantage of compression and varchar columns to save as much space as possible. The space saved may allow you to build at least one badly needed index).

3. Denormalizing would be my last resort if space allows.

Art

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