Forum

 
ForumForumDiscussionsDiscussionsSQLBI Methodolo...SQLBI Methodolo...SQLBI Methodology more questions/commentsSQLBI Methodology more questions/comments
Previous Previous
 
Next Next
New Post
 5/27/2009 11:54 PM
 

Hi,

I saw your presentation at the European PASS conference. It was definitely one of the highlights of the conference.
And it is great that you bring the topic “methodology” into focus again. After reading your papers too I still have some
questions. Here is my main concern:

Let’s see if I’ve got it right. Your presentation in Neuss suggested to me, that the “Inmon aspect” of your methodology
doesn’t consist in his top-down approach, but in adding a relational level (of “indirection”) to the architecture – the
data warehouse – which contains the complete truth. Contrary to Inmon this relational Layer is incrementally build,
thus rescuing Kimballs agility (at least partly). And contrary to Kimball the Data Marts are merely derived from this “single
point of truth”. They don’t constitute it.

The problem with new levels of indirection is of course the price you have to pay. Views come pretty cheap. So that
makes sense. The additional relational layer is way more expensive (the mirror DB too). So my question is, what exactly
do we gain and what do we pay.

The main advantage you seem to promise is reduction of complexity, complexity of the main data model, the data model
which represents the “single point of truth” within the whole of the BI solution. Is this correct? And do you think, there
is a systematic reason why a relational model should capture complex scenarios better than a dimensional one?

Two more questions/comments:

  • Why do you bind Data Marts to departments? Kimball (and I strongly agree) binds them to processes. In fact I have the
    impression, that some of the complications you mention start right here.
  • You say „a BI solution is completely based and driven by data, not by user requirements“ (another member of this forum
    already mentioned this statement). I agree, you cannot base it on (articulated) requirements. If the users already know all
    the appropriate questions and requirements, they probably don’t need the BI solution in the first place. But I guess just
    looking at the plain data won’t help you either. The ERP will probably produce tons of data, nobody cares about. And some
    of the most interesting stuff may be hidden in some obscure system nobody tells you about. Isn’t knowledge of the costumers
    (core) processes a much better driver?

 OK, I guess that’s enough for one post.

   

Thanks for your great contributions to the community

 

 

 

 

Christoph

 

 

New Post
 6/1/2009 11:39 AM
 

Cristoph,

 

first of all, thank you for your comments: they are a great value!

In general, consider that SQLBI approach is long-term oriented. The need for an higher number of levels in the architecture (and for a way to decouple them as much as we can) borns from our real-world experience. The complexity of relationships in real-world data sometimes does require complex relational models, with a lot of relationships between entities, most of them optionals and not mandatory. Trying to put these relationships in a "flat" star schema often requires a simplification and possibly the loss of some information, just because the user don't need them at that particular point in time. However, we don't want to lose precious data for future analysis (and future requirements!) and we also don't want to force a dimensional model to be too much complex, including data for future requirements. Too much complex models are not used at all, in our experience.

For these reasons, we say that a "single poin of truth" can be better expressed using a relational model than a dimensional one. You have to think in a long-term perspective, anyway.

Data Mart for Departments or Processes - sincerily, with SQLBI we create Data Mart when they are needed, following the requirements that can be different each time. I think there is not so much importance here in discussing where the Data Marts have to come from. Oftentimes, we had to build completely different data marts using the same source data just because of the different representation of data required by end users. In that specific case, we don't want to spend time discussing if the difference were about departments (they were different) or processes (they could be considered different, but in reality we were analyzing exactly the same data, but from two differents points of view - could we say they were part of different processes?).

Finally, yes, the knowledge of customers processes is a key factor for the success of a BI solution (or of "any" solution!) and we can't simply define our Data Warehouse looking only at the OLTP data sources. In general, you always have to find the right balance. The point we wanted to highlight is that you cannot define your Data Warehouse model basing only on user requirements of the final reports they want to build. At the same time, you cannot design that model basing only on existing structures of source data. You need the knowledge of both and, of course, the knowledge of processes that makes that data meaningful!

I really appreciate this kind of discussions. Thanks again for your feedback.

Marco

Previous Previous
 
Next Next
ForumForumDiscussionsDiscussionsSQLBI Methodolo...SQLBI Methodolo...SQLBI Methodology more questions/commentsSQLBI Methodology more questions/comments