Design and development of algorithms for identifying termination of triggers in active databases

Abstract

An active database system is a conventional database system extended with a facility for managing triggers (or active rules). Active database systems can react to the occurrence of some predefined events automatically. In many applications, active rules may interact in complex and sometimes unpredictable ways, thus possibly yielding infinite rule executions by triggering each other indefinitely causing non-termination. Almost all of the work to date on termination analysis for active databases relies on checking that a directed graph called the |triggering graph| for a set of rules is acyclic. This paper presents new algorithms for detecting termination/non-termination of rule execution using triggering graph. These algorithms detect termination/non-termination among the rules in a triggering graph without any limitation on the number of rules.