Data Modeling Essentials, Third Edition 3rd Edition

This is a book structured in three parts.Part one addresses the basics in data modeling. The need for modeling when designing databases is explained and alternative modeling approaches such as ER and UML discussed. The ER approach is the one developed in detail. Usual data modeling concepts such as Classes, Generalization, Specialization, Attributes, Relationships, Normalization, etc. are introduced and explained here.The chapter on primary keys discusses the different candidates for key selection and provides valuable guidelines for choosing them. A short chapter on extensions and alternatives to the ER approach, in which the UML notation is addressed, ends part one.Part two is about the organization of the data modeling task and the data modeling cycle, from the requirements phase to the conceptual, then to the logical data models. Each phase is developed in detail and illustrated through sensibly chosen examples. Thorough pro and con argumentation is provided. How data modeling fits in the project perspective, the role of different actors involved - the business specialist, the data modeler, the database designer, etc. - and project organization matters are addressed too.Part three addresses advanced normalization - BCNF, 4NF and 5NF - in terms that are as simple and accessible as reasonably possible, compared to other data modeling books that pretend achieving the same level of completeness. Modeling of time-dependent data and business rules, barely avoidable in real-life databases, is also covered.A short chapter of 21 pages on data warehouses and data marts addresses OLAP data modeling. That's plenty, as the book mainly focuses on OLTP data modeling. Another short chapter whose purpose is to provide a quick overview of enterprise data modeling and of data management aspects in general ends the book.This is a well-written, well-illustrated and complete book that I liked very much. I highly recommend it. Without doubt, it's an outstanding book written by capable and rigorous authors.It's also a thick and dense book that beginners should precede by shorter, introductory readings.

I'm a data architect at a small tech company, now working mostly with data warehouses, and was looking to refresh my data modeling skills. After reading Amazon reviews about different modeling books I chose this one and have not been disappointed. Rather than reading it from front to back, I read the first couple hundred pages and am now skipping around and reading sections that interest me.The book is logically put together, and has a very detailed contents index, which makes finding relevant information easy. The sections I have read not only explain the theory but also give good examples putting the theory into practice. However, they sometimes seem to place too much emphasis on a theoretical approach that would never be used in the real world.Overall I find this book very useful and have marked it up with sticky notes for sections I'll revisit for my next database modeling design.

I've been a data modeler for many years and think this book, plus the Kimball Data Warehouse Toolkit and Len Silverston's three book set on reference data models are the best.I was afraid that Hadoop and NoSQL killed OLTP and OLAP data modelling, but it seems many startups have created a real mess data model wise and that data modelling as a discipline is making a comeback. Poor data models lead to excessive hacks and patchwork code.

There is a great deal of excellent material contained within the corpus of this text. But to be fair, much of the material is also far less than first rate. In particular, the first and last chapters were excellent. These provided, respectively, an overview of the discipline, with some philosophical underpinnings, and a very excellent, sensible, and readable review of Enterprise Data Modeling strategy. On the other side of the ledger, the chapter on subtypes was very weak, as were several others.One supposes that such a mixed bag of goods results from two writers attempting together to produce a common message. In this case, that attempt fails. And what we see instead are two markedly different forms of outreach and instruction. I didn't care to discover which author wrote which section. But the results were obvious. It is up, in my view, to the authors and the editors to sort this out in hopes of producing a better quality result in the future.We would recommend the book, in a qualified sense, to the practictioner. The excellence does outweigh, and prevail over, the mediocrity, even though it is, at times, a "close run thing", as someone famously said. And we would hope also that the authors and editors would take these candid comments to heart. God bless.

Data Modeling Essentials scores a perfect 10 on the subject of data modeling. The topic is difficult and yet this book makes it possible for any aspiring data modeler to be effective if he or she is willing to put in the time. The book is easy to read. The examples are very good. The topic is challenging but with this tutorial you have a chance to be a good data modeler. Even if you are a developer or DBA not responsible for data modeling you will definitely be glad you purchased this book. Just do it.