Archives of the TeradataForum

Message Posted: Wed, 22 Feb 2006 @ 08:44:10 GMT

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

I would concur with the rest of the folks on the full table scan for partial list of columns that make PI, that's one key difference which beginers in teradata overlook.

However I think there's a possibility in the case of multi column secondary indexes in the sense that optimizer may choose to do a index subtable scan (again that's a full table scan on the subtable, but since subtable is smaller, it's fast) for partial list of columns that makes the index. which again I believe depends on the fact that you have proper statisitics on the index and the individual columns.

any thoughts folks ? :)

Joseph Vinish D'silva

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