Algorithms For Joint Task Scheduling And Computational Offloading In Mobile Cloud Computing

Many applications have been developed to meet the various needs of people's lives.Most tasks of applications are more complex and heavier,as the development of applications.However,the computing power and resources of mobile devices are usually limited,which makes it difficult for users to meet real-time and efficient requirements.Currently,mobile cloud computing technology is the common method to solve the above problems.However,not all tasks of the application are suitable to be offloaded to the cloud server.When a task is offloaded to cloud server,the mobile device needs to consume the corresponding energy and time.In order to minimize the energy consumption of mobile devices and meet the real-time requirements of users,it is necessary to design a combination of task offloading and scheduling.For the combination of task offloading and scheduling issues,current researches mainly use IBM's optimizer CPLEX to solve it.But this methods require a lot of memory and computing resources and the running time of the program is very long.Moreover,with regard to the problem of joint scheduling and offloading solved by the optimizer,the heterogeneity of the task is not considered.Therefore,this method is not suitable for practical problems in mobile cloud computing.For the problem about joint task scheduling and offloading in the mobile cloud computing,we present a mathematical model that is more suitable for physical problems according to existing researches.The mathematical model considers the heterogeneity of tasks in an application and we propose a heuristic algorithm to solve the problems studied in this paper.The heuristic algorithm is mainly composed of the process of task offloading and scheduling.The offloading of the process requires the process of calling the scheduling in our heuristic algorithm.The task offloading process uses a greedy strategy to minimize the energy consumption of mobile devices.The scheduling process ensures that all tasks of the application are finished before its deadline,and the scheduling algorithm is based on the priority of tasks in the application.Experiments prove that the heuristic algorithm compared with the traditional methods can obtain the approximate optimal solution of the problem in a short time.Wealso proposed a simulated annealing algorithm to optimize the solution of the heuristic algorithm.We found that the solution of the heuristic algorithm is very close to the solution of the simulated annealing algorithm.It is further illustrated that our heuristic algorithm is efficient and feasible.