Novel Sustainable and Heterogeneous Offloading Management Techniques in Proactive Cloudlets

Abstract

Cloudlet-based mobile cloud offloading is an emerging technology designed to augment mobile elements by migrating resource-hungry components to adjacent local resource pooling. However, the Cloudlet resources are usually limited in terms of the computing utility, storage and network bandwidth. In this scenario, the remote Cloud infrastructure can provide additional computing and storage utility during run-time; the heterogeneous offloading methods for different mobile applications and diverse offloading resources complicate the Cloudlet-based offloading and resource allocation process. As a result, a considerable amount of delay is caused by setting up the execution environment, communication overhead and waiting in the queue, which significantly downgrade the QoS and the usability of such systems. In this paper, we propose a novel hybrid offloading model to solve the heterogeneous resource-constraint offloading issues in the Cloudlet, concerning the offloading energy and execution efficiency. A queue-based offloading framework is developed to formulate and analyze the mixed migration-based and partition-based offloading behaviors in the Cloudlet. The execution and energy-aware heterogeneous offloading resource allocation problem is formalized, and a Particle Swarm Optimization heuristic solution is presented. A SARIMA-based load prediction model is designed in the Cloudlet to achieve fine-grain proactive resource allocation. Experimental results reveal that the proposed framework can effectively reduce the offloading energy cost and execution time, compared to currently existing solutions.