摘要
针对ad hoc云中的任务卸载问题,设计一种多目标任务卸载决策模型。综合考虑任务完成时间、能耗和额外开销进行卸载决策,并选取簇头节点作为集中控制器进行合理的任务分配。提出一种融合遗传算法和蚁群算法的任务卸载算法,利用遗传算法的快速搜索能力得到可行解,将其作为蚁群算法的初始信息素,再利用蚁群算法的正反馈机制实现对任务分配方案的精确求解。仿真结果表明,该算法与随机任务分配算法、异构感知任务分配算法和遗传算法相比,能有效降低任务完成时间和能量消耗。
Aiming at the task offloading problem in ad hoc cloud,this paper designs a multi-objective task offloading decision model.The task completion time,energy consumption and overhead are comprehensively considered to make the offloading decision and select the cluster head node as the centralized controller for reasonable task assignment.This paper also proposes a task offloading algorithm based on genetic algorithm and ant colony algorithm.We utilized the fast search capability of genetic algorithm to obtain the feasible solution,and then used it as the initial pheromone of ant colony algorithm.According to the positive feedback mechanism of ant colony algorithm,we obtained the task assignment scheme accurately.The simulation results show that the proposed algorithm can effectively reduce task completion time and energy consumption compared with random task assignment algorithm,heterogeneity-aware task allocation algorithm and genetic algorithm.
作者
余思东
黄欣
赵志刚
Yu Sidong;Huang Xin;Zhao Zhigang(Department of Information and Electromechanical Engineering,Guangxi Agriculture Vocational and Technical College,Nanning 530007,Guangxi,China;College of Computer and Electronics Information,Guangxi University,Nanning 530004,Guangxi,China)
出处
《计算机应用与软件》
北大核心
2020年第11期185-191,199,共8页
Computer Applications and Software
基金
国家自然科学基金项目(61363067)
广西高校中青年教师科研基础能力提升项目(2019KY1408)。
关键词
ad
hoc云
多目标
任务卸载
遗传算法
蚁群算法
ad hoc cloud
Multi-target
Task offloading
Genetic algorithm
Ant colony algorithm
