Mastering Data Warehouse Design: Relational and Dimensional Techniques

by Claudia Imhoff, Jonathan G. Geiger, Nicholas Galemmo

A cutting-edge response to Ralph Kimball's challenge to the data warehouse community that answers some tough questions about the effectiveness of the relational approach to data warehousing Written by one of the best-known exponents of the Bill Inmon approach to data warehousing Addresses head-on the tough issues raised by Kimball and explains how to choose the best modeling technique for solving common data warehouse design problems Weighs the pros...

• Format: Paperback
• Language: English
• ISBN: 0471324213
• ISBN13: 9780471324218
• Release Date: August 2003
• Publisher: Wiley
• Length: 456 Pages
• Weight: 1.55 lbs.
• Dimensions: 1.0" x 7.4" x 9.4"

Customer Reviews

Excellent perspective and practical guidance

I have many books on the topic but this one has a special spot on my shelf. It toggles b/w high and low level concepts effortlessly to cover DW design and architecture in a way that provides guidance for a wide range of scenarios. The book addresses key cutting, DW audits and controls, hierarchy explosion tables, etl/data modeling/change mgt.. the list does go on. There are suggestions for sequencing out the activities involved in DW construction. This is a refreshing text that blends the large schools of DW thought and in the process comes up with all sorts of new insights that can be used in the trenches.

Excellent Reference for Data Warehouse Design

This book is a must read for anyone exploring or undertaking a data warehouse. It discusses the many issues regarding a data warehouse and balances the views authored by both Bill Inmon and Ralph Kimball. As the former publisher of DM Review magazine I have worked with many of the experts in data warehousing and have found that Claudia Imhoff is one of the best in the industry. Claudia is excellent at articulating and presenting the key issues and critical information that is essential to gain a firm understanding for data warehousing. She presents the theory behind the concepts which has been lost by many in the industry. At DM Review, I tried to get Ralph Kimball and Bill Inmon to debate the many issues behind data warehousing and data marts with little success. Bill Inmon was open to the debate but Ralph was not. Claudia is able to look at both viewpoints and separate what really matters between them. The major premise that still holds true today is that a data warehouse is essential to any organization that is trying to make tactical and strategic decisions. It should be built in an iterative fashion and should be the foundation for the creation of data marts. This book should be a key reference for anyone building a data warehouse.

Integration Of Inmon & Kimball Thinking

"Mastering Data Warehouse Design" is an excellent book to help readers understand how to take maximum advantage of the strengths of diverse approaches associated with Bill Inmon and Ralph Kimball.The main reason I bought a copy of this book, even before it arrived in bookstores, was that I was leading a team to figure out how to merge Inmon and Kimball views for data modelling standards.We had already developed a DW architecture using Inmon's approach, with its associated relational/ERD method, but believed that it lacked rigour in the area of data marts. We also reviewed Kimball's books, and acknowledged the strengths of his dimensional modelling approaches, but were concerned that it lacked rigour for the diversity of analytical requirements in the manufacturing environment, e.g. data exploration/mining on a massive scale. We were struggling to figure out to combine the best of both - and then we discovered the imminent release of "Mastering Data Warehouse Design". After checking the Table of Contents on the publisher's web site, we had the book couriered directly from the publishers warehouse because it would not be available in local bookstores fast enough to meet our work schedule.Chapter 1 has an impressive 'sound bite' version of Inmon's DW architecture thinking, but extended to include broader Business Intelligence concepts. Chapter 2 does a commendable job of explaining a tiered approach to data models, e.g. subject area model, business model, Operational system model, DW model. At first, this chapter was confusing because we had just finished a rigourous definition of data modelling standards, using more conventional terminology, e.g. logical/entity model, physical/table model. So the book's terminology didn't seem to fit in with our thinking. But after re-reading it, we realized that it added value in forcing us to look at the whole issue of modelling from a deliverables or outcomes perspective, rather than a modelling process perspective.Chapter 4 discusses how to develop a DW data model. The content outlines the sequence or steps involved in developing a DW data model, and it's rare that I've been able to find as good coverage of the topic as I found in this chapter. Chapters 5 - 11 cover topics like keys, modelling time/hierarchies/transactions, with some solid content on how to model for on-going business change and how to maintain the tiered models. However, I'm not fully conversant with some of these topics, so am not in a good position to evaluate their content.Chapter 12 has a very good discussion on how to deal with a proliferation of legacy data marts, and strategies for migrating to a central DW that feeds a variety of data marts. It also introduces Chapter 13 which has a classic discussion on comparing the relational and dimensional modelling approaches - including the best discussion I've ever seen on the strengths and weaknesses of each approach. While our team didn't buy into all this chapter's points, the clear logical

A top class encyclopedia!

An excellent compendium of Data Warehousing, Modeling, and Management processes. It is a detailed practical-guide for IT implementers and a terrific framework for Architectes to optimize productivity. Reality based discussion of trade-offs in fast changing market, enterprise, and customer will come in handy in everyday decisions. Even though I live in the DW-BI World, I found kernels of truth often compromized in favor of saving time, and appreciated a refreshing view of pros and cons for doing it right, whether from start or while upgrading / integrating. You will find yourself going back to many sections to share with your staff. A great read and terrific reference for your DW-BI reading. It sits on my reference shelf with many dogeared pages and underlined sections.

All in one Guide, Text-Book, and handy Reference

Experienced and new to DW-BI professionals will find this work useful. It shows how to cure the ills from past systems and streamlie going forward. It packs not just technical methodologies and structure, but also practical guidelines for real-time implementation. You will be surprized and delighted by how much can be improved even in a well-run shop.