Constraint-Aware Complex Event Pattern Detection over Streams

In this paper, we propose a framework for constraint-aware pattern detection over event streams. Given the constraint of the input streams, our proposed framework on the fly checks the query satisfiability / unsatisfiability using a lightweight reasoning mechanism. Based on the constraint specified in the input stream, we are able to adjust the processing strategy dynamically, by producing early feedbacks, releasing unnecessary system resources and terminating corresponding pattern monitor, thus effectively decreasing the resource consumption and expediting the system response on certain situations. Our experimental study illustrates the significant performance improvement achieved by the constraint-aware pattern detection framework with little overhead.