OLAP Solutions: Building Multidimensional Information Systems [With CDROM]

A book which gives you a comprehensive theoretical foundation together with useful, pratical techniques to design Olap solutions.In the first half of the book the author explains you all you need to think clearly in N dimensions, in the second half he guides you through the design of a rather complex application, stepping into the purchasing process, the material inventory analysis, sales & marketing, and activity-based management. He teaches you using maieutics, Socrates' way of teaching, via a dialogue between two designers. This way, a lot of questions that usually arise during real-life development are examined and solved.Whilst the examples in the first edition of this book were based on TM1, in this second edition they are written using a vendor-independent language, called LC (Located contents), which provides you with a solid, logically based foundation to multidimensional design.The vendor languages, like those used by Microsoft Analysis Services (MDX), Applix TM1, Oracle Express and Hyperion Essbase, are kind of Plato's shadows in the cavern of this ideal language (which is sort of a tautology for MS Plato...), so if you start thinking in LC, you'll be able to work with all of these products.Ideal complements to this book, if you use Microsoft Analysis Server, are "MDX Solutions" by George Spofford, and "Microsoft Olap Solutions", by Thomsen (again!), Spofford and Chase.

Even though I had been familiar with OLAP concepts before I bought this book, I enjoyed reading it and recommend it to anyone seriously interested in OLAP. You will learn the history of OLAP and why traditional tools like SQL and spreadsheets are not very well suited for analysis that multidimensional tools are capable of performing. There are two comprehensive chapters on basic multidimensional features like dimensions, hierarchies, attributes and so on for those who are not familiar with them yet. There are more chapters describing some advanced features like joining cubes, visualization, formulas. Sparsity problems are taken into consideration as well as storage options for multidimensional data. There are several business cases for OLAP application, but, unfortunately, as other reviewers have already noticed the software which goes with the book does not work properly, so I just skimmed throuth the chapters where the use of the software was described. For me it was not the main part because I had already had some experience with OLAP tools before. The author destinguishes clearly between data warehousing and OLAP which some people sometimes fail to do. I liked also the last chapter where the author gives some guidelines on how to evaliate OLAP tools, which may be very helpful for those who have not made their minds yet about what tool to use in their project. Although there is some outdated material in the book, it without a doubt deserves its five stars.

If you're not doing MSOLAP, then this is pretty much (as of the date of this writing) the only practical choice for an OLAP book. It's a good book, very clear explanations of some concepts I was having trouble with (ie, difference between 'generation' and 'level' when every vendor treats these words differently). The descriptions of multi-dimensionalism and how to train your brain to think non-two-dimensionally were just what I needed. I did not install the software in the back of the book, so I can't comment on that.

As some one who makes a living doing OLAP development, I thought the Author does a great job explaining the similarities and differences between OLAP and relational analysis. It is important to note that OLAP analysis does not replace relational analysis - it enhances it.

This is a book that everyone dealing with OLAP must read (if is possible, the first one to be read on the subject). Precise conceptualization of the OLAP showing differences between its capabilities and other alternatives for surfing information in datawarehouses. If you read this book in conjunction with Ralph Kimball's books on Data Warehouse you will get the best information available.