A Hybrid Approach for Improving Elasticity in the Cloud Computing Environment

Elasticity is considered as one of the most important features that distinguishes cloud computing from other distributed system approaches. This feature takes into account the fact that the resource allocation process is considered as a process that can be implemented dynamically. Providing an efficient solution for improving elasticity will be useful for both providers and users of cloud computing services. Using the proposed solution in this paper, providers will be able to evaluate and improve the quality of their services, and increase their qualitative or quantitative advantage in competing with other competitors. In this paper, we present a hybrid solution for improving elasticity through using buffer management and centralized elastic management. Buffer managment controls the input queue of the request and the elastic management by using the reinforcement learning controls the elasticity of the system. Finally, we evaluate the effectiveness of our approach under three real workload traces, namely, Yahoo Cluster, Wikipedia, and Google Cluster workload traces. The experimental results show that the proposed approach reduces the response time by up to 15.2%, and increases the resource utilization by up to 13.2 % and the elasticity by up to 19.8 % compared with the CTMC and ControCity approaches.