摘要：For cloud providers, workload prediction is a challenging task due toirregular incoming workloads from users. Accurate workload prediction is essentialfor scheduling the resources to the cloud applications. Thus, in this paper, the authorspropose a predictive cloud workload management framework to estimate the neededresources in advance based on a hybrid approach, which is a combination of animproved Long Short-Term Memory (LSTM) network and a multilayer perceptronnetwork. By improving the traditional LSTM architecture by using opposition-baseddifferential evolution algorithm and dropout technique on recurrent connectionwithout memory loss, the proposed approach has the ability to perform a betterprediction process. A novel hybrid predictive approach is aiming at enhancing theprediction performance of the cloud workload. Finally, the authors measure theproposed approach's effectiveness under benchmark data sets of NASA andSaskatchewan servers. The experimental results proved that the proposed approachoutperforms the other conventional methods.
