A New Approach for Event Triggering Probability Estimation in Active Database Systems to Rule Scheduling Improvement

Active database systems (ADBS) can react to the occurrence of some predefined events automatically. Reactive behavior of ADBS is organized by a collection of active rules. One of the most important modules of ADBS is the rule manager. The main responsibility of the rule manager is triggering, buffering, firing and selecting (scheduling) rules. Rule scheduling approach has considerable impact on performance and efficiency of ADBS. In this paper, we propose a new approach for improving the rule scheduling in ADBS. We first introduce a framework to compare and evaluate existing rule scheduling approaches. In this framework, five evaluation criteria have been proposed. Average Response Time, Response Time Variance, Throughput, Time Overhead per Transaction and CPU Utilization. Existing approaches have been evaluated by using this framework and the approach which has the most positive impact on performance and efficiency of ADBS has been selected by analyzing the weaknesses and strengths of existing approaches. Then, to improve the selected rule scheduling approach, we developed an Event Triggering Probability Estimation algorithm and integrated this algorithm to selected rule scheduling approach. Results of experiments show that the new proposed algorithm increases the positive impact of selected rule scheduling approach on performance of ADBS.