Archives of the TeradataForum

Message Posted: Tue, 24 Feb 2004 @ 10:45:41 GMT

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

Hi Nitin,

Check the distribution of the answer set based around the primary index of the target table. The target table is defined as SET but with no UNIQUE indexes on it, this is not bad but it means that the DBMS has to do duplicate row checks. With lots of rows for the same primary index data value this can be very expensive. This will typically be the biggest difference between your select and the insert/select. Ignoring the dup row check, I'd also check the distribution anyway, this may not show up in the select but may start to cause problems in the insert/select.

Cheers,

Dave

Ward Analytics Ltd: Information in motion ( www.ward-analytics.com )

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