Object Databases (251-0355-00)

Lecturer: Dr. Alexandre de Spindler

Wednesday 09.15-11.00, CAB G 59 (Lecture)
Wednesday 11.10-11.55, CAB G 59 (Exercises)

Course Description

When object-oriented programming languages became the status quo in application engineering, the impedance mismatch between relational databases and object-oriented applications grew into a serious problem and object-oriented databases emerged also. While purely object-oriented databases were able to overcome the drawbacks of the relational model, they often lacked support for database management. For example, few systems implemented a standardised common query language and rather relied on the programming language to retrieve and manipulate data. These limitations and the fact that relational databases also provided some support for object-orientation have restricted the success of object-oriented databases. However, object-oriented database management systems have recently become popular again as embedded databases used in the automotive or aerospace industry.

The course examines the principles and techniques of providing data management in object-oriented programming environments. It is divided into three parts that cover the road from simple object persistence, to object-oriented database management systems and to advanced data management services. In the first part, object serialisation and object-relational mapping frameworks will be introduced. Using the example of the open-source project db4o, the utilisation, architecture and functionality of a simple object-oriented database is discussed. The second part of the course is dedicated to advanced topics such as industry standards and solutions for object data management as well as storage and index technologies. Additionally, advanced data management services such as version models for temporal and engineering databases as well as for software configuration are discussed. In the third and last part of the course, an object-oriented data model that features a clear separation of typing and classification is presented. Together with the model, its implementation in terms of an object-oriented database management system is discussed also. Finally, an extension of this data model is presented that allows context-aware data to be managed.

Assistant

Tilmann Zäschke