摘要
云计算环境中的任务调度问题一直是云计算研究的重点,任务调度的目的寻找最优的任务调度策略,以高效地完成计算任务。针对云计算环境下资源规模庞大、异构性的特点,为了克服传统调度算法存在的缺点,提出一种基于改进自适应人工鱼群算法的任务调度算法。该算法以任务总执行时间作为目标函数,在迭代过程中动态自适应的调整人工鱼的视野和步长,同时对觅食行为进行改进,加快算法的收敛速度,避免算法陷入局部最优,以此提高任务调度的性能。通过在Cloud Sim平台进行仿真对比实验,相较于其他智能寻优算法,该算法在任务执行时间和收敛速度上都有明显的优势,是一种有效的任务调度算法。
Tasks scheduling is an important issue to be resolved in cloud computing research. The purpose of task scheduling is to find the best optimal scheduling scheme to compute tasks efficiently. Focusing on the characteristics of resources under large scale heterogeneous in cloud computing environment, and to overcome the shortcut of the existing task scheduling algorithm, a task scheduling algorithm based on improved self-adaptive artificial fish swarm algorithm was proposed. The algorithm used the task execution time as objective function, adjusted artificial fish vision and step dynamically in iterative procedure, and improved prey behavior, to speed up the convergence rate, reduce the probability of local optimum, and improve the performance of task scheduling. Through the simulation experiment on CloudSim platform, the algorithm has obvious advantages in the task execution time and convergence speed compared to other intelligent optimization algorithms, it is an efficient task scheduling algorithm.
出处
《电子设计工程》
2017年第6期14-18,共5页
Electronic Design Engineering
基金
陕西省网络计算与安全技术重点实验室项目(15JS078)
西安市科技计划资助项目(CXY1518(1))
关键词
云计算
任务调度
人工鱼群算法
自适应
cloud computing
task schedule
Artificial Fish Swarm Algorithm （AFSA）
self-adaptive
