Browse Books

Reviewer: Sergio Ilarri

This is an excellent book that provides an in-depth overview of all issues related to distributed data management. I was eager to read it, since not many books focus on distributed databases, and since the best reference in the field [1] and the previous edition of this book [2] are more than ten years old. I was not disappointed. This updated edition presents some topics in more detail (such as database integration, multidatabase query processing, and data replication) and introduces new topics (peer-to-peer (P2P) data management, Web data management, stream data management, and cloud computing). Chapter 1, "Introduction,"? and chapter 2, "Background,"? explain basic concepts of distributed databases and the fundamentals of relational databases and computer networks, respectively. Chapters 3 and 4 focus on design. On the one hand, chapter 3, "Distributed Database Design,"? describes the top-down approach for the design of distributed databases. It covers (primary and derived) horizontal fragmentation, vertical fragmentation, hybrid fragmentation, fragment allocation, and the use of a data directory. On the other hand, chapter 4, "Database Integration,"? focuses on the bottom-up approach, which is appropriate for multidatabase systems. It addresses the problems of schema matching, schema integration, schema mapping, and data cleaning. Chapter 5, "Data and Access Control,"? focuses on semantic data control: view management, data security (data protection and access control), and semantic integrity control. The authors consider each topic, first from the perspective of centralized systems, and then from extended to distributed databases. Chapters 6 through 9 deal with query processing. The first three chapters consider tightly coupled homogeneous distributed database systems, and the last one focuses on multidatabase systems. Chapter 6, "Overview of Query Processing,"? introduces the problem of query processing. Chapter 7, "Query Decomposition and Data Localization,"? describes the first two stages in the generic layering scheme for distributed query processing. Chapter 8, "Optimization of Distributed Queries,"? deals with query optimization, and focuses mostly on join operations. Finally, chapter 9, "Multidatabase Query Processing,"? explains the mediator/wrapper architecture for multidatabase query processing. Chapters 10 to 13 focus on transaction management and data replication. Chapter 10, "Introduction to Transaction Management,"? presents basic concepts. Chapter 11, "Distributed Concurrency Control,"? assumes no failures and no replication, and presents concurrency "control algorithms that provide the isolation and consistency properties of transactions."? Chapter 12, "Distributed DBMS Reliability,"? acknowledges the possibility of failures by extending the previous techniques to guarantee the atomicity and durability of transactions. Finally, chapter 13, "Data Replication,"? presents techniques to maintain the consistency of replicated data items. Chapter 14, "Parallel Database Systems,"? presents the different parallel database system architectures, the problem of parallel data placement and query processing, load balancing, and database clusters. Chapter 15, "Distributed Object Database Management,"? focuses on fundamental object database technology in the context of distributed databases. It introduces basic concepts of object models, and discusses issues regarding object distribution design, architectural issues, object management, distributed object storage, object query processing, and transaction management. Chapter 16, "Peer-to-Peer Data Management,"? deals with data management for "modern"? P2P systems (characterized by high distribution, heterogeneity, and volatility), including types of P2P networks, schema mappings, querying (top- k queries, joins, and range queries), and replica consistency. Chapter 17, "Web Data Management,"? discusses Web graph management, Web search, Web querying, and distributed Extensible Markup Language (XML) processing. Finally, chapter 18, "Current Issues: Streaming Data and Cloud Computing,"¿ presents some aspects of data stream management (although most existing works propose a centralized architecture) and cloud data management. Each chapter (except for the first introductory chapter) ends with a conclusion that provides a nice summary and additional reflections, as well as bibliographic notes. Most chapters also include exercises. The book ends with an extensive list of references and an index. The publisher's Web site provides accompanying slides for "instructors who adopt the book."¿ One of the few minor negative points may be that the book contains a number of typographical errors and editing problems. Regarding the contents, more exercises on distributed database design, as well as answers to some of the proposed exercises, would have been useful. Mobile databases and semantic techniques for data access/management on the Web are two interesting topics that the book does not significantly cover. It only makes a few mentions of commercial database systems, since the book focuses on the principles, rather than existing systems. Instructors of advanced database courses could use this book as a textbook. It would also interest researchers on topics related to distributed data management. I highly recommend this excellent book. Online Computing Reviews Service