基于移动云计算的计算迁移能效算法

Energy-efficient computation offloading algorithm in mobile cloud computing

移动云计算可以通过应用任务的计算迁移降低执行延时和改善移动设备能效,但面对多云站点选择时,迁移决策是NP问题。针对该问题,提出一种能效计算迁移算法。为了实现截止期限和预算约束下执行时间与代价的多目标优化,算法将优化过程分解为三步进行。首先根据用户对时间与代价参数的偏好,设计一种CTTPO算法对应用进行分割,生成迁移模块(云端站点执行)和非迁移模块(移动设备执行);然后为了实现云端多站点间的迁移模块调度,设计一种基于教与学最优化方法的MTS算法,进而产生效率最优的应用调度解;最后设计一种基于动态电压缩放方法的ESM算法,通过多站点的性能缩放进一步降低应用执行能耗。通过两种随机应用结构图进行了仿真实验,实验结果证明,该算法在执行效率、执行代价以及执行能耗上要优于对比算法。

Mobile cloud computing can use the computation offloading of application tasks to improve the energy efficiency of mobile devices and reduce the execution delay of applications.However,when facing the multiple service sites selection from cloud,the offloading decision is a NP problem.In order to address this problem,this paper proposed an energy-efficient computation offloading algorithm.To complete the multi-objective optimization of the execution time and the execution cost under the deadline and budget constraints,this algorithm divided the optimization process into three steps.Firstly,it proposed CTTPO algorithm to divide the application according to the user preferences with respect to time and cost parameters,which could give the offloadable(executed on cloud sites)and non-offloadable modules(executed on mobile device)of the application.Se-cond,it proposed a MTS algorithm based on teaching-learning optimization method to schedule offloaded modules among multiple cloud sites and generate the time optimized efficient scheduling.Finally,it proposed an ESM algorithm based on dynamic voltage scale method to further reduce the execution energy consumption through the performance scale among multiple sites.It constructed two types of random application structure graphs to run simulation experiments.The experimental results show that this algorithm performs better than the baseline algorithm on the execution time,the execution cost and the execution energy.