Archives of the TeradataForum

Message Posted: Wed, 04 Apr 2001 @ 16:18:57 GMT


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

Subj:		Re: Primary keys in Fact tables for Star Schemas

From:		Patrick Logan

Dave Ahnell wrote:

We do not have the time to redesign the data structures into a more normalized design, realizing that this works the best with Teradata.

I am new to Teradata and am scheduled for courses in a couple of weeks. Meanwhile I've been reading manuals in my spare(!) time.

I have yet to find a definitive reference for the often heard claim, above. What exactly does "works best" mean for Teradata? I understand that Teradata helps a 3NF design "work better" than on some other database technologies.

But how does one quantify "3NF works better than dimensional" for Teradata? What is the formal explanation and where is the quantified evidence? I am not simply curious, we are beginning data modeling and this not only affects "performance" but development effort, front-end tool support, maintenance & evolution, etc.

Thanks,

Pat