Mastering Data Modeling: A User-Driven Approach

Mastering Data Modeling: A User-Driven Approach
by John Carlis & Joseph Maguire
Copyright 2001
Addison-Wesley

Find out more about this book, or purchase it, from Amazon.com

Probably the most important thing you can do to create a fast and scalable SQL Server-based application, and one that meets the needs of the user's it serves, is to perform proper data modeling and database design at the earliest stages of the development cycle. As DBA or developers, we are often tasked with the job of data modeling. Unfortunately, we often haven't been trained properly in data modeling, and we design applications that are less than optimal.

Data modeling is generally one of those subjects you learn in college as a Computer Science major. But as luck would have it, many DBAs and developers don't have a degree in Computer Science. Their knowledge and experience has been gained from the real world, and data modeling is not one of the skills you just pick up.

If you have been given the job of creating a new SQL Server-based application from scratch, including the task of data modeling, and don't know where to start, you may want to pick up a copy of the book, Mastering Data Modeling. This book is designed to start you from a blank slate, and guide you into becoming a master data modeler. Taking very much a textbook approach, the book provides you both with a strong theoretical and practical knowledge of how to model data.

The book describes what is called Logical Data Structure (LDS), which is a data modeling notation methodology used for data modeling. The book shows you how, through a step-by-step process, to collect, model, and document data structures and flows.

The book includes these topics:

• Learning good data modeling habits.

• Learning how to read Logical Data Structure (LDS) with sentences.

• The vocabulary of LDS.

• Data visualization techniques.

• Learn how to converse with users in order to learn about their data.

• Learn how to master the shapes used with LDS.

• Using one-entity, no-relationship shapes.

• Using one-attribute shapes.

• Using two-entity shapes.

• Using shapes with more than two entities.

• Using shapes with reflexive relationships.

• Learning LDS rules.

• Learning how to come up with proper entity names.

• Using official names.

• How to label links.

• How to document an LDS.

• Learn how to script the flow of the data.

• How to control the evolution of the data model.

• Learning about data constraints.

• Learning about data-modeling notation.

• How LDS fits into the relational model.

If these topics sound like a foreign language, it some ways they are. Much of data modeling revolves about taking "real-world" data and modeling it for a relational database. This requires a modeling language of its own, which in this case, is LDS.

If you are new to data modeling, you will want to read this book more than once. In your first reading, speed through it in order to get the "big picture". Then in your second reading, take your time, working through all of the examples. Ideally, try to do this when actually starting a data modeling project. This way, you will be better able to apply what you are learning to your project at hand, and things will make much more sense to you.

I recommend this book for anyone who wants to learn more about data modeling, especially DBAs and developers tasked with this job for the first time.