Abstract:

Current data-driven applications operate in a global context with shared data being modified by several applications. Database management systems (DBMSs) provide mechanisms for sharing large amounts of data and manipulating them by many concurrent users in a consistent manner. However, they were not designed to support monitoring of subsets of relations by multiple users when they are being changed or manipulated by several users. In many applications, it is not only necessary to monitor these subsets but also to visualize the changes in some way. For example, in one of the applications, the track relation is being updated with respect to tracks (or objects) and their positions by multiple sources. Many users monitor subsets of these tracks as needed for their use (e,g, different regions).

The objective of this thesis is to overcome the limitations of current DBMSs with respect to change monitoring and notification. We present an approach that leverages active technology to enable dynamic change monitoring of database relations in an efficient and transparent manner. In particular, we extend a mediator based Active Technology system, the ECA Agent and discuss its architecture, design and implementation details.