A time series and reduction-based model for modeling and QoS prediction of service compositions

摘要

Web services are emerging as a major technology for deploying automated interactions between distributedrnand heterogeneous applications. The accurate prediction of their quality of service (QoS) is importantrnbecause their users rely on it to decide whether they meet the QoS requirement. The existing studies of QoSrnprediction usually assume that QoS of service activities follows certain distributions. These distributionsrnare used as static model inputs into stochastic process models to obtain analytical QoS results. Instead, wernconsider the QoS activities to be fluctuating and introduce a dynamic framework to predict the runtime QoSrnby employing an Autoregressive Moving Average Model and QoS reduction rules. In the case study of arnreal-world composite service sample, a comparison between existing approaches and the proposed one isrnpresented, and results suggest that the proposed one achieves higher prediction accuracy.