Dynamic resource allocation in hybrid mobile cloud computing for data-intensive applications

Mobile cloud computing is a platform that has been used to overcome the challenges of mobile computing. However, emerging data-intensive applications, such as face recognition and natural language processing, imposes more challenges on mobile cloud computing platforms because of high bandwidth cost and data location issues. To overcome these challenges, this paper proposes a dynamic resource allocation model to schedule data-intensive applications on integrated computation resource environment composed of mobile devices, cloudlets and public cloud which we refer as hybrid mobile cloud computing (hybrid-MCC). The allocation process is based on a system model taking into account different parameters related to the application structure, data size and network configuration. We conducted real experiments on the implemented system to evaluate the performance of the proposed technique. Results demonstrate the ability of the proposed technique to generate an adaptive resource allocation in response to the variation on application data size and network bandwidth. Results reveal that the proposed technique improves the execution time for data-intensive applications by an average of 78% and saves the mobile energy consumption by an average of 87% in compared to using only a mobile device while the monetary cost increased only 11% due to using cloud resources and mobile communication