Virtualized Resource Provisioning for Hybrid Cloud Computing Services Using Optimal Data Analytics

Abstract Resource Provisioning is the important method to offer cloud as services to the user. Nowadays cloud computing has multiple access and application levels. We need a hybrid and virtualized method to provide efficient resource provisioning method. In existing resources can be offered based on user request or availability. But current scenario we need any time access privilege based services offering systems. Also data analytics is another factor to analyse the data and make effective decision making approach. We propose an efficient virtualized dynamic resource provisioning method for cloud application and virtualized optimizer for topological cloud management. We prepare optimal data analytics method to predict the accuracy and performance of cloud access and resources. In this paper automated machine learning technique is applied to measure the temporal features and queuing results. In this case loads are distributed to server and optimize the cloud using service level agreement. Our research method can provide the result of time service analysis, accuracy index, performance factors and optimality value. Also our method can test over provisioning and energy consumption. We use CloudSim for simulating cloud environment and TensorFlow for testing accuracy using Deep belief network optimization.