移动云计算中的一种任务联合执行策略

发布时间：2018-02-14 22:36

本文关键词： 云计算移动互联网应用迁移能效遗传算法　出处：《计算机学报》2017年02期 　论文类型：期刊论文

【摘要】：针对移动用户将应用迁移至云端处理会引起大量的数据传输导致高能耗的问题,提出了一种任务联合执行策略(Task Collaborative Execution Policy,TCEP).首先,在云端和移动终端联合执行移动应用的前提下,将应用考虑为一系列的串行任务,根据任务的计算负荷、输入和输出数据量,把云端与移动终端联合移动应用的优化问题建模为最小化移动终端的能耗问题,并得出结论该优化问题属于NPC(Nondeterministic Polynomial Complete)问题.接着,按移动终端需向云端迁移任务的次数来划分迁移策略集,并利用串行任务仅能一个接一个地执行的特点,给出了一次迁移最优特性.然后,通过对比串行任务与染色体的相似点,采用遗传算法来处理文中优化问题,并在简单遗传算法(Simple Genetic Algorithm,SGA)的基础上,利用一次迁移最优特性来设计交叉操作和变异操作,以便进一步提高算法性能.最后,通过仿真验证了所提策略及算法的性能,仿真结果表明,改进后的遗传算法具有良好的收敛性能,能够保证新个体具有仅向云端迁移一次的特性,与现有方法相比,所提策略可有效地减少搜索最优解的运算时间,能在满足应用执行时间要求的同时最小化移动终端的能耗.
[Abstract]:In order to solve the problem that migration of mobile users to cloud processing will cause a large amount of data transmission and lead to high energy consumption, a joint task execution strategy, Task Collaborative Execution Policy, is proposed. Firstly, under the premise of joint execution of mobile applications by cloud and mobile terminals, a task joint execution strategy is proposed. The application is considered as a series of serial tasks. According to the computing load, input and output data of the task, the optimization problem of the combined mobile application of cloud and mobile terminal is modeled as the problem of minimizing the energy consumption of the mobile terminal. It is concluded that the optimization problem belongs to the NPC(Nondeterministic Polynomial complete problem. Then, the migration policy set is divided according to the number of tasks that mobile terminals need to migrate to the cloud, and the serial tasks can only be executed one by one. In this paper, the optimal property of one time migration is given. Then, by comparing the similarity points between serial tasks and chromosomes, genetic algorithm is used to deal with the optimization problem in this paper, and on the basis of simple genetic algorithm (SGA), a simple genetic algorithm (SGA) is used to solve the optimization problem. In order to further improve the performance of the algorithm, the crossover operation and mutation operation are designed by using the one-time migration optimality. Finally, the proposed strategy and the performance of the algorithm are verified by simulation, and the simulation results show that, The improved genetic algorithm has good convergence performance and can ensure that the new individual has only one migration to the cloud. Compared with the existing methods, the proposed strategy can effectively reduce the computational time of searching the optimal solution. It can minimize the energy consumption of mobile terminal while meeting the requirement of application execution time.
【作者单位】：中国联合网络通信有限公司湖南省分公司;中南大学信息安全与大数据研究院;北京邮电大学信息与通信工程学院;北京邮电大学计算机学院;
【基金】：国家自然科学基金(61173017,61563038) 国家“八六三”高技术研究发展计划项目基金(2014AA01A701) 内蒙古自治区高等学校创新团队发展计划支持项目(NMGIRT-A1609) 2013年度湖南省教育厅科学研究项目(13C613)资助~~
【分类号】：TP3;TN929.5

【参考文献】