Using Partial Differencing for Efficient Monitoring of Deferred Complex Rule Conditions

This paper presents a difference calculus for determining changes to rule conditions in an active DBMS. The calculus has been used for implementing an algorithm to efficiently monitor rules with complex conditions. The calculus is based on partial differencing of queries derived from rule conditions. For each rule condition several partially differentiated queries are generated that each considers changes to a single base relation or view that the condition depends on. The calculus considers both insertions and deletions. The algorithm is optimized for deferred rule condition monitoring in transactions with few updates. The calculus allows us to optimize both space and time. Space optimization is achieved since the calculus and the algorithm does not presuppose materialization of monitored conditions to find its previous state. This is achieved by using a breadth-first, bottom-up propagation algorithm and by calculating previous states by doing a logical rollback. Time optimization is...

• Type: Conference
• Year: 1996
• Where: ICDE
• Authors: Martin Sköld, Tore Risch
• Keywords: Bottom-up Propagation Algorithm, Database, Deferred Rule Condition, ICDE 1996, Rule Conditions