Volume 18, No. 6, 2021
An Energy Efficient And Deadline Constrained Heuristic Algorithm For Dynamic Offloading Of Tasks In Mobile Cloud Computing Environment
Ramesh Babu A , Dr. Niraj Upadhayaya
Abstract
In MCC environments, mobile applications run based on battery power of smart hand held devices. It the presence of constrained resources, dynamic offloading of tasks to public cloud is essential for mobile applications to meet task deadlines and enhance performance of the system. When all the tasks are executed locally, it leads to high resource consumption and inability in meeting deadlines. Similarly, when all the tasks are offloaded to cloud, it results in more execution time which may sometimes even lead to missing deadlines. Therefore, it is essential to strike balance between with optimal offloading decisions selectively. Towards this end, in this paper, we proposed an algorithm named Energy Efficient and Deadline Constrained Heuristic Algorithm for Dynamic Offloading (EEDCHA-DO). It considers the problem of deadline constrained tasks offloaded to cloud based on execution time and energy consumption in presence of varied bandwidth of wireless network. The algorithm performs desired computations dynamically for each task and the problem of whether to offload or not to offload the task is solved optimally. The algorithm is implemented in MATLAB and the simulation results revealed that EEDCHA-DO is efficient in task offloading when compared with its predecessors such as LAC and LC. The proposed algorithm achieved minimal energy consumption while meeting the deadline of tasks with its optimized offloading decisions.
Pages: 4302-4315
Keywords: MCC, cloud computing, computational offloading, local execution, cloud execution