Adaptive Job Load Balancing Scheme on Mobile Cloud Computing with Collaborative Architecture

Abstract

The adaptive mobile resource offloading (AMRO) proposed in this paper is a load balancing scheme for processing large-scale jobs using mobile resources without a cloud server. AMRO is applied in a mobile cloud computing environment based on collaborative architecture. A load balancing scheme with efficient job division and optimized job allocation is needed because the resources for mobile devices will not always be provided consistently in this environment. Therefore, a job load balancing scheme is proposed that considers personal usage patterns and the dynamic resource state of the mobile devices. The delay time for computer job processing is minimized through dynamic job reallocation and adaptive job allocation in the disability state that occurs due to unexpected problems and to excessive job allocations by the mobile devices providing the resources for the mobile cloud computing. In order to validate the proposed load balancing scheme, an adaptive mobile resource management without cloud server (AMRM) protocol was designed and implemented, and the improved processing speed was verified in comparison with the existing offloading method. The improved job processing speed in the mobile cloud environment is demonstrated through job allocation based on AMRM and by taking into consideration the idle resources of the mobile devices. Furthermore, the resource waste of the mobile devices is minimized through adaptive offloading and consideration of both insufficient and idle resources.