Customer Reviews

Building the Data Warehouse (3rd Edition)

5 star:		(6)
4 star:		(4)
3 star:		(3)
2 star:		(1)
1 star:		(2)

 

 

Bill's books are good conceptual books aimed at explaining the Business Benefits and Overall Technical concepts that are fundamental to data warehousing. This book is a classic that people serious about Business Intelligence should read. HOWEVER - Inmon's technical details are light and sometimes generalized, whereas Kimball's books tend to be more nuts-n-bolts and include usable examples. I RECOMMEND (a) read Inmon for how BI environments should be built under ideal situations, (b) read Kimball (warehouse toolkit) for how to do an actual project. For Project Management and Infrastucture issues, read Kimball's "Warehouse Lifecycle Toolkit". (c) Read Inmon's Corp Information Factory to understand the BI architecture most companies try to build (whether they know it or not!), and (d) STOP wasting time comparing Inmon to Kimball - their paradigms are complimentary, not contradictory.
CHECK OUT "Data Warehousing Fundamentals" by Paulraj Ponniah - it covers concepts by Inmon, Kimball and other leading thinkers and is very useful when -planning projects, -working with management to explaing BI environments.

The two approaches, Inmon and Kimball are complimentary. Kimball is not "BS" and "stupid" as stated by one reviewer. In fact, it is the PRACTICAL techniques and elegant SIMPLICITY that makes Kimball's work so valuable. Inmon is strong on theory, while Kimball is strong on the day-to-day tools required to build a useful and understandable decision support environment. This reviewer has observed many huge "pure Inmon" projects fail (and none observed succeeded) while all pure Kimball projects observed have thrived and exceeded expectations. Strong words, to be sure, but accurate to the reality observed by this reviewer. While the failures might be attributed to the particular technical teams and management groups, many of those same groups did succeed with a dimensional approach.

If you want to debate academics, read both authors and have at it. I have to actually deliver... on time or early and on budget or below. Thus, my rating reflects that reality. Give me Kimball. The techniques have worked for me (and teams of all sizes) since 1986 at many Fortune 100 and 500 clients.

Inmon provides a great perspective on the issues and on how to think about information management for decision support and this book provides some of his latest thinking on the topic. Kimball provides an excellent and proven way to get there in the real world.

I know the Inmon vs. Kimball thing has turned into something of a feud. Having owned the first and second edition of this particular book (the first edition is now more useful than ever hollowed out with a knife to hide jewelry) and having browsed through the third edition that a colleague purchased, I must say that the major change since the obscure first edition is that Inmon has now taken a much more defensive tone now that Kimball is out there presenting clear, concise and useful real-world concepts and implementation techniques and gathering followers. I also own Inmon's "Using the Data Warehouse", "Data Warehousing and the Zachman Framework", and "Exploration Warehousing" and the question that comes to mind most after reading through these books is why did I waste the money. The reading is difficult, not so much because the subject matter is difficult, but more because it seems to be dry theory and has little advice on how to actually design a real working data warehouse. When later, Kimball came out with his breakthrough book: "The Data Warehouse Toolkit", it all seemed conceptually so much simpler and straightforward.

If you work for a large corporation which has millions of $ to spend on DW projects, maybe you should look at this book and even consider some of the ideas that it contains.
But if you need to develop a data warehouse using limited resources and within a certain timeframe, your time will better used reading other books, because following the Inmon approach will lead you to an unnecessary complicated and expensive design.
I found that the arguments used by Inmon to demonstrate the limits of the dimensional approach are not convincing at all. For example, at page 142 he says "Because there is a different data structure for each data mart, making any data mart into a data warehouse doesn't make sense."
Having personally implemented several data warehouses using the "conformed dimensions" approach, I can guarantee that it worked and produced a very elegant and clean data model.

I was a little skeptical about what was new to cover in a third edition, but the author really opened my eyes to new ideas that he's pioneering--like integrating data warehouses with ERP systems and using near-line storage techniques for better performance. His ideas about online techniques for customer support were also provocative. Bottom line is that this remains the classic overview of data warehousing. It is still the best and most comprehensive introduction around--and now also the most up-to-date.

As other reviers have mentioned, this book covers wide, but not that deep. Like many books, it gives advice so generic as to be useless, but an easy to understand overview has its importance too, so perhaps this is a good place to start.

So you will see the basic problems of combining data from different sources into one place, but not necessarily how to do it. Very little is dedicated to the issues around moving that data or ETL.

This book has many chapters, many of which may not apply to your situation.

This book is a good introduction to data warehousing. However, the style is remarkably bad. It is very repetitious, poorly organized overall, occasionally self-contradictory, and jam-packed with cartoon-like line drawings that seldom add clarity to the discussion. The book easily could be reduced from 500 pages to 200 pages without losing any information, and with an improvement in readability.

In my opinion Kimball vs Inmon is not a war of religion, they both have pro and cons in different situation. I believe people seriously interested in BI should read the book and take several advices from it. There are situation in which an Inmon design is good, there are some in which it isn't BUT to be able to judge you need to know both an be able to decide based on the customer's necessity.
The pictures in the book are really ugly and useless, nevertheless the concepts are clearly stated and easily understandable. If I have to say something wrong about it is that you need to already know what a BI system is in order to get the most out of the book. It is not for newcomers but it is definetely a good book on BI.

Data warehouse development relies on a scientific substrate, but usual target reader is not a computer scientist, nor a programmer, rather an advanced user who has to present dwh projects to management. So most books are mainly business-level presentations, and this book is not an exception. It contains some technical issues, more in the form of hints and tips than fully developed solutions, that have nowadays become common and wide-known features. Most books about dwh, OLAP and even mining look like as they have been written some years ago (typically late ninety) in order to present to a wide audience those new disciplines, and then submitted to further and slight revisions to work out new editions. In my experience, user community's skill has grown up, but many publications have not been able to keep its pace.
However, though this does not seem not a mandatory step in a dwh basic education, it still retains all its historical importance.

A little over fifteen years ago, I was working for the then great, and now deceased, Chicago area consulting firm of Greenbrier and Russel. On their behalf, I was often called upon to speak publicly on matters relating to logical data modeling and relational database design. One of these occasions was a meeting of the Milwaukee chapter of the Data Administration Management Association. When I spoke before these wonderful folks, the speaker following me was one Bill Inmon. How could I have known then how close was I to one of the real future "rock stars" of the Data Warehousing World? (I must also report that some folks said I gave the better presentation that day. But that had everything to do with the excellent presentation graphics put together by the outstanding staff at G&R. Inmon, obviously, still lacks such artistic, professional support.)

In any case, Inmon's book, now reviewed, is really quite a good introduction to the important, and still emerging, topic of corporate data warehousing. It is, as some other reviewers have averred, a bit light technically. However, it does provide something of a "soup to nuts" introduction to the topic. I was particularly struck by how nicely Bill dealt with the evident controversy still extant between what is styled as his data warehousing approach versus that of Ralph Kimball. Inmon respectfully cites Kimball in the corpus of this text. In so doing, he simply, and I'd judge fairly, illustrates the difference between the approaches, allowing his readers to reach their own conclusions. In the sometimes acrimonious world of technical debate, I found this approach refreshing and praise worthy.

As a practitioner of many years in the discipline, I can recommend Inmon's book as an excellent and comprehensive introduction to the important and still emerging topic. God bless.