Archives of the TeradataForum

Message Posted: Fri, 02 Jul 2004 @ 11:54:53 GMT

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

Its a duplicate across all amps and then crutially a sort in the same step. This means that Teradata can use a merge join as the duplicated spool is now in join columns sequence.

With a duplicate across all amps and then a product join, there is no sort in the duplication step as it would not be necessary.

Regards,

Martin

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