A Service Selection Method of Non-critical Task in Parallel Structure

Abstract

Aiming at the neglected selection for non-critical task in parallel structure, this paper provides a method of services selection based on grey relational analysis. It regards the execution time of critical task as the upper limit for the execution time of non-critical task, uses the execution time of non-critical task as the reference series, and takes other QoS criteria as compared series. And the grey relational grades represent the relational degree between the QoS criteria and the execution time; the grey relational grades are reduced to the QoS criteriapsilas weights, then evaluating the weighted sum of criteria in order to choose the optimized candidate Web QoS service of parallel structure. Though the selection sacrifices the execution time of non-critical task, it has no influence on the execution time of parallel structure, and improves other QoS criteria simultaneously. The instance analysis shows that the method is a composite service QoS optimization technique which does not affect the execution time of composite services and obtains the better service quality at the cost of sacrificing the execution time of non-critical task.