Re: Joining fact tables?

From: Vikas Agnihotri <vikasa_at_writeme.com>
Date: Mon, 25 Oct 1999 12:43:10 GMT
Message-ID: <7v1j8u$st$1@nnrp1.deja.com>

In article <7utldv$1ime$1_at_node17.cwnet.frontiernet.net>, "Clark Morrow" <cemorrow_at_frontiernet.net> wrote:

> To get the full performance advantage of a dimensional datawarehouse,
you
> need to have a RDBMS that performs star optimizations and you need to
have a
> query tool that performs multi-pass SQL. Of the two, it is more
important
> to have a tool that performs multi-pass SQL.

I see. Any recommendations?

> It is a good thing to combine multiple fact tables. Indeed it is a
> requirement.

Hm. Most people seem to warn against this. From what I know, the thinking is that "one fact table per datamart". If you are joining 2 fact tables, you are combining apples and oranges (different subject areas) i.e. your initial design was flawed.

Would you care to share any examples/design criteria to back up this assertion that joining fact tables is "good".

Thanks,
Vikas

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Mon Oct 25 1999 - 07:43:10 CDT