A Learning-based Approach for improving of resource provisioning in cloud computing environment

The rapid development of cloud computing has led to the proliferation of various data centers around the world, that has increased the number of data centers, the number of resources with similar performance but different specifications. Cloud services come with new concepts such as elasticity and scalability. One of the most important differences between traditional services and cloud services is their elasticity. In this paper, a method to improve the resource provisioning for the cloud computing environment is presented, that has four phases of monitor, analysis, plan and execution. In the monitor phase, data is received and the analysis phase, data is pre-processed. The decision phase, that is the most important phase, uses the Bayesian learning technique to decide on the provision of cloud resources. Finally, the result of the decision phase is applied to the resources by the execution phase. The novelty of this paper is the use of Bayesian technique and its combination with the methods used in the analysis phase. The performance results of the proposed method show an increase in elasticity of 5.05% and an increase in elastic accuracy of 6.59% and a scaling rate of 4.31% compared to the compared methods