Many to Many Relationship

[Waynest]

Hi

I am new to OLAP and have been dropped in at the deep end with the 2005 version

My first objective with this is to get some sales aggregations out of a simple cube, but I've been wrestling with it for days now and Im not getting far. I will spell out my simple table structure & maybe someone will be kind enough to walk me through this.

Each of our customers has a unique entry in a customer table. Each customer may be a member of many 'customer groups' and each 'customer group' may contain many customers.

So I have a customer table, a customer-group table, and a group table. I think this is how MS intends us to design this kind of thing? I have a single fact table 'sales' which I want to be able to aggregate by customer and/or group

Im tired of fumbling around in the dsv and cube designers could someone help get me started with this pls.

Thanks

 

[unclejimbob]

The best way to get started is to grab yourself a book that quickly gives practical advice and demonstrations of processes using real(ish)-world problems.

I strongly recommend you borrow/buy/beg/steal 'The Data Warehouse Toolkit (2nd edition)', read through to the 3rd chapter - roughly 66 pages - and you will have a very good understanding as to the basics of dimensional modelling, certainly enough to get you through this first problem.

good luck

 

[johnherman]

There is also a free tutorial on dimensional modeling at this web address, but I also recommend the Kimball book

http://freedatawarehouse.com/tutorials/default.aspx

-------------------------
The trouble with doing something right the first time is that nobody appreciates how difficult it was - Steven Wright

 

[Waynest]

Thanks for replies.

I've managed to work out how to incorporate the relationship using the Dimension Usage Tab and have some reports based on my cube coming out of Reporting Services now.

I've got Teo Lachev's new book but its as dry as a witch's tit. Maybe they all are