Research Of Active Database Rules Based On Graph Method

Recently active database has been the focus of the database research field. Active database technology is studied for the auto-maintenance data integrality and coherence and to satisfy the requirement of real information process. It is the landmark of database theory development .It makes database system can monitor and analyze database state inside and outside automatically. It makes real time correspondence possibly .It is the theory and technology base of solving database design problems.Alexander Aiken pointed out there were three aspects of rule analysis of static analysis method: termination, confluence and observable determinism. Is rule processing guaranteed to terminate after any set of changes to the database in any state? The processing of active database rules trigger other rules of the same rule, the triggering of rules may be non-termination. Can the execution order of nonprioritized rules make any difference in the final database state? To determine confluence, at first we should consider the rule commutativity. Satisfy confluence requirement and no non-termination graph, the rules in R are confluence.Determining termination of an active rule set is one of the problems that affecting the design of active database directly. No cycles in TG is a sufficient condition of the termination of rule sets, but it is not a necessary condition. Through study of cyclic active database rule sets and analysis of triggering graph and activation graph, we present the concept of united formula and the algorithm to determine termination of a rule set corresponding to a cyclic triggering graph. We also present the time complexity analysis of the algorithm.We determine confluence by analyzing the rule's commutativity. With the method of rules execution graph, we analyze the confluence of rules set and corresponding algorithm is given as well.