On Transaction Boundaries in Active Databases: A Performance Perspective

The workload of an active DBMS consists of two types of activities: externally generated tasks submitted by users, and rule management tasks caused by the triggering of rules stored in the knowledge component of the system. Most design proposals for active DBMS’s assume that an external task should be combined with all the resulting rule management tasks into a single transaction. There is no compelling reason for this assumption, however; the semantics of the rules can be used to divide the workload into transactions in a number of different ways. In this paper, we describe a performance model designed for studying active DBMS performance issues. We present the results of simulation experiments that study system performance as a function of transaction boundary semantics for varying levels of data contention, rule complexity, and data sharing between externally submitted tasks and rule management tasks. Our results demonstrate that the way in which transaction boundaries are imposed can have a major impact on the performance of an active DBMS. We therefore conclude that this aspect of rule semantics must be carefully considered at the time that rules are specified.