Modern Database Systems: The Object Model, Interoperability, and Beyond

Read an Excerpt

PREFACE:

This book brings together in a single volume a discussion of many of the issues that must be addressed in order to solve two technical challenges facing database system users (end users and application developers) in the 1990s:

• The transition from relational database technology to postrelational technology
• Provision for the coexistence and interoperability of old and new database technologies

Since the 1960s, the technology for managing data has evolved from file systems to hierarchical, to network, to relational. The 1980s were a gestation period for the new-generation technology. There was a flurry of activity to develop database systems that could meet the following requirements:

• Support an object-oriented data model
• Provide an architecture that easily can accommodate various functional extensions
• Support the management of, among others, temporal data (history and versioning), spatial data, multimedia data, long-duration transactions, imprecise data, and rules

This activity was fueled by the emergence of database applications that today's commercial relational database systems cannot support to the desired level of productivity in application development and run-time performance. Today, however, there is almost enough knowledge to develop commercially viable next-generation systems. Thus, the first challenge to meet in the 1990s is the transition from relational database technology to postrelational database technology.

However, next-generation systems so far have not replaced older-generation systems. This has led to the coexistence of database systems of all generations.Therefore, the second database challenge of the 1990s is to bring about interoperability of the database and file systems installed world-wide during the past three decades.

In the 1980s, database vendors began offering gateways that allowed relational database applications to retrieve data from nonrelational database systems or file systems. At the same time, database researchers worked to lay a foundation for building multidatabase systems --systems providing a single database view over various independently developed database and file systems. Such systems will allow users to access various databases and files in one, uniform database language without the users' being aware of the heterogeneity of the underlying systems. Here, too, there is almost enough knowledge to develop commercially viable systems.

This book, rather than being a mere collection of already published papers, is a compendium of contributions that are largely original. I specifically solicited the chapters from leading database researchers, database developers, and industry experts, and edited each chapter for technical accuracy and uniform style of presentation. Space limitations prevented me from giving a more in-depth treatment to each topic, but I believe each has been covered sufficiently to give the reader a good grounding on the subject. Each chapter includes a comprehensive overview of the issue covered, proposed solutions to problems, and directions for further research and development. I hope this book will help readers understand the current status of the database field and where it should go.

I enjoyed working with the expert contributors, and thank them for their submissions. I also want to thank Helen Goldstein of Addison-Wesley Publishing Company, Inc., and Nhora Cortes-Comerer of ACM Press Books, both of whom worked with me from conception through final production of this book. Finally, I want to thank Janelle Larson and Catherine Richardson for painstakingly converting the chapters, which were received from the contributors in many different formats, into a single uniform style in Framemaker. Their work reduced the production schedule by as much as six months.

W.K.

Read More