Agile Data Warehouse Design: Collaborative Dimensional Modeling, from Whiteboard to Star Schema

by Lawrence Corr, Jim Stagnitto

Agile Data Warehouse Design is a step-by-step guide for capturing data warehousing / business intelligence (DW/BI) requirements and turning them into high performance dimensional models in the most direct by modelstorming (data modeling + brainstorming) with BI stakeholders. The book describes BEAM, an agile approach to dimensional modeling, for improving communication between data warehouse designers, BI stakeholders and the whole DW/BI development team. BEAM provides tools and techniques that will encourage DW/BI designers and developers to move away from their keyboards and entity relationship based tools and model interactively with their colleagues. The result is everyone thinks dimensionally from the outset! Developers understand how to efficiently implement dimensional modeling solutions. Business stakeholders feel ownership of the data warehouse they have created, and can already imagine how they will use it to answer their business questions. Within this book, you will

Genres: Technology, Programming, Nonfiction, Reference, Technical, Computer Science, Bisexual

328 pages, Paperback

First published November 24, 2011

Reviews

[Hillary · Rating 4 out of 5]

Very well written, easy to read style, excellent real world examples. The main thing was not just what to do, but why. I learn best when I truly understand and this book gave me what I was looking for.

[Hunter Johnson · Rating 4 out of 5]

Excellent overview, and supported by a worthwhile website with downloads. A few items were presented as things to do without enough reason why. But I took several good ideas from it to implement immediately.

[Malvin · Rating 4 out of 5]

Using BEAM, Business Event Analysis and Modelling as the agile modelling method for designing dimensional data warehouse, the author delivers a comprehensive guide, addressing how to capture the 7Ws - who, when, what, where, why, how many, how and why, working collaboratively and iteratively with the customer, to avoid BDUF (Big Design Up Front) and BRUF (Big Requirements Up Front). The book includes design patterns and best practices on how to model each of Ws covering the various types of Fact tables, historical, multivalued and "junk" dimensions, support for different time zones across different languages and considerations for performance, with examples, highlighting typical design problems and suggested solutions. Overall, the content is more suitable for non-beginners. Experienced BI developers will be familiar with some of the content but will more likely to appreciate the recommended agile approach to delivering a data warehouse.

[Daniel Galassi · Rating 4 out of 5]

I think the authors put together two important aspects of BI solution design, a design methodology and a good introduction to dimensional modelling. I found the methodology easy to follow and formulated in a way that can be used with business users.
The methodology is probably easier to implement in greenfield implementations, since "reverse engineering" a mid-sized data warehouse and generate the necessary supporting elements may be a time consuming activity. That said, the methodological approach of BEAM* would help consulting teams tremendously.
Big thumbs up!

[John Anderson · Rating 4 out of 5]

Very good system. If you have the discipline to follow this you can create a functional data warehouse. More geared for a consultant though, hard to see how you would go through this for only one company.

[Yasmine Dumouchel · Rating 4 out of 5]

Very well written and actionable. Presents a pragmatic and structured methodology to gathering requirements in DW/BI context at the same time as using dimensional modelling. It is a more practical version of Kimball's book.

[Guy Fawkes · Rating 4 out of 5]

The book consists of two parts:
- The first is describing the collaborative requirements gathering that involves stakeholders and uses Agile methodology. Among others it's also a good way to make your stakeholders more interested in data. While it's interesting I don't think is that realistic to follow it in most companies due to resources constraints.
- The second part dives into specific design challenges and shares patterns and techniques you can use to tackle them. It's pure gold! It follows dimensional modeling and it's paired great with Kimball. I would single out the hierarchy modelling which is approached with great depth and detail.