Workload Forecasting with Hybrid Stochastic Configuration Networks in Clouds

With their fast development and deployment, the cloud data center providing a large number of service which has become the most import service of Internet.. In spite of numerous benefits, their providers face some challenging issues. Workload forecasting plays a crucial role in addressing them. Accuracy and fast learning are the key performances. Its consistent efforts have been made for their improvement. This work proposes an integrated forecasting method that combines Savitzky-Golay filtering and wavelet decomposition with Stochastic Configuration Networks to get the workload forcast in the next period. In this study, we adopt Savitzky-Golay filtering to smoothing a task number sequence, and then the smoothed series is decomposed into multiple components by wavelet decomposition. Based on them, integrated prediction model is for the first time established and the statistical characteristics of trend and detailed components can be well characterized. The results of our study demonstrate that the proposed method has better performance than some typical methods.