Resource Prediction Using Hybrid Double Seasonal Decomposition in Fog Integrated Cloud

Time series prediction has many real-world applications such as weather forecasting, electric utility forecasting, load forecasting in the cloud and many more applications. In this paper, we propose a hybrid model to predict the workload in fog integrated cloud architecture. The prediction of workload is divided into two domains: the long term and the other short term. The underlying system models and time-series data generating processes are generally complex for these applications and the models for these systems are usually not known a priori. The underlying motivation for using CTREE and CART is the ability to accurately forecast time series data when the process requests of the types are nonlinear, non-stationary and not well-defined a-priori, however, ARIMA methodology also we have used to compare the results and used to forecast the trend part. In this paper, we propose different sets of prediction models, including CART and CTREE and compare the result with other statistical-based predictions. We have used the Wikimedia data set for the prediction model and compared it with other states of the art. CTREE and CART have also been proven to outperform other linear techniques including AR, MA, and ARIMA prediction techniques. Simulation results in our proposed model showed good accuracy, leading to efficiency and dynamic resource utilization with good quality of service(QoS).