The Performance Prediction of Cloud Service via JOGM(1,1) Model

Abstract

With the rapid development of cloud computing, more and more cloud services are provided to users. Faced with multiple cloud services, how to scientifically predict and assess the performance of cloud service is an imperative task. To cope with the challenge, an prediction scheme via JOGM(1,1) model is proposed based on grey system theory, in which the performance of cloud service is quantified as the response time. Thus cloud users and the third-party institutes of cloud service performance evaluation can predict and assess the performance of cloud service as accurate as possible. In return, this will contribute to the cloud service provider selection for better performance on behalf of cloud consumers. The simulation results show that the proposed scheme has higher prediction precision compared with classic GM(1,1) model and weighted moving average model, which verifies the effectiveness of the proposed prediction scheme and the feasibility of forecasting the performance of cloud service employing grey system theory.