Browse Books

Reviewer: Michael Hanspeter Böhlen

Parts of this book, which was designed as a text for an introductory course in database systems, are also valuable as a reference for database experts. In particular, the discussion of duplicates, constraints, recursion, Datalog, and selected aspects of SQL2 and SQL3 goes beyond what is offered in standard database textbooks. Among the salient features of the book are its comprehensive coverage of SQL, including integrity constraints, assertions, triggers, bags, and recursion in SQL3; the material on the design of databases using both the entity-relationship (ER) model and the ODMG standard ODL; and a discussion of object-oriented and object-relational database programming, including the ODMG standard OQL and the object-oriented features of SQL3. Compared to standard database textbooks, this book represents quite a departure in introducing database systems. Rather than covering a wide range of areas relevant to database systems, the authors only cover the material most useful to the majority of computer science students. This means that database systems are approached from the points of view of database designers, users, and application programmers. Throughout, examples and exercises are used extensively. Often, theoretical explanations are kept short, and one or two examples are used instead. This approach improves the book's accessibility, even when it discusses advanced topics. Exercises are provided with every section, making the book suitable for self-study. Solutions to many of the exercises are accessible over the Internet. Equally useful are the reminder boxes, which are used frequently to point out important concepts. The introductory chapter delimits the scope of the book. The focus is on database design and programming. Implementation aspects, such as query processing, transaction processing, and storage structures, are not elaborated on. Chapter 2 introduces database modeling. For modeling purposes, the authors use both the well-established ER model and the ODMG standard ODL. The running example is modeled using each approach in turn. This allows readers to easily compare the ER model and ODL. Chapters 3, 4, and 5 cover the relational data model and SQL. Beginning with the mapping from the conceptual model to relations, these three chapters cover the basic aspects of relational databases, including functional dependencies, normal forms, multivalued dependencies, the relational algebra, and SQL. The introductory material is completed by more advanced discussions of Datalog, recursion, and bags. Advanced aspects of SQL are discussed in the next two chapters. Chapter 6 offers a comprehensive and insightful discussion of constraints and triggers in SQL. Chapter 7 discusses system aspects of SQL with which every SQL programmer needs to be familiar. This includes the embedding of SQL into a host language, SQL transactions, schemas and catalogs, and security and authorization issues. Chapter 8 again adopts a broader view. It discusses object-oriented databases by comparing selected aspects of SQL3 and OQL. Ullman and Widom make the motivation for database systems one of their primary goals. They achieve this goal by structuring the book according to the needs of database designers and users and by discussing issues that are highly relevant to commercial database systems but not usually covered in standard database textbooks. I can warmly recommend the book both as an introductory text on database systems and as a reference on many practical aspects of SQL. Readers interested in database implementation techniques, such as physical database design, indexing techniques, query processing and optimization, and transaction processing, will need to supplement the book with a second textbook.