Adaptive Processing for Continuous Query over Data Stream

Abstract

Stream applications such as sensor data processing, financial tickers and Internet traffic analysis require that information, naturally, occur as a stream of data values. Due to a late and out-of-order arrival of infinite, unbound and multiple input streams, processing continuous queries over them may lead to producing an incorrect answer or delaying query execution. Hence to minimize this waiting time, previous works have used timeout technique without considering the frequency of timeouts. It results in decreasing the accuracy of query execution results, since the more the frequency of timeouts, the more the loss of data. We propose an AP-STO method using StB that stores operator’s state and a window time-out method based on the waiting time for the next tuple by resetting the size of a window according to the frequency of timeouts. It reduces a data lost rate and increases the tuples output-rate. We compare AP-STO method with an existing method and use output-rate and response time as criteria for performance evaluation. Our proposed method shows a substantial improvement in system performance in terms of the accuracy of query execution and the increment of tuples output-rate per a query due to the reduction in loss rate of data.