摘要
针对云计算的高能耗问题,从系统级节能角度,提出一种节能的资源调度算法。首先,建立云计算的两级资源调度模型;综合考虑主机的工作、空闲和休眠等多种状态建立能耗模型,并用多功能计量插座加以验证。然后,提出基于遗传算法的最小能耗资源调度算法(minimum energy consumption based on genetic algorithm,MECGA),根据云任务的服务质量(quality of service,QoS)需求产生初始种群,以系统能耗最小为调度目标设计适应度函数,并根据染色体适应度的正态分布函数和种群的进化代数设计遗传算子。仿真结果表明,所提算法能够有效降低系统总能耗、缩短任务完成时间。
To solve the high energy consumption in cloud computing, from the system-level energy saving, an energy efficient resource scheduling algorithm in cloud computing environment is proposed. First of all, the model of two level resource scheduling in cloud environment is established. Considering the different states of resources, such as sleep, idle and working, the energy consumption estimation is modeled, and it is verified by a multifunction meter. After that, a minimum energy consumption resource scheduling algorithm based on ge- netic algorithm （MECGA） is proposed. In MECGA, the initial population is produced based on the quality of service （QoS） requirements of cloud tasks, and the fitness function is designed according to the scheduling ob- jective. Furthermore, the normal distribution function of the fitness and the evolutional generation of chromo somes are used to design the genetic operator. The simulation results show that the proposed algorithm has a better performance in both task completion time and energy consumption.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2013年第11期2416-2423,共8页
Systems Engineering and Electronics
基金
国家自然科学基金(61071093)
江苏省研究生科研创新计划项目(CXZZ12_0483
CXLX12_0481)
江苏省科技支撑计划(BE2012849)
江苏高校优势学科建设工程(yx002001)资助课题
关键词
云计算
能耗模型
资源调度
遗传算法
CloudSim平台
cloud computing
energy consumption model
resource scheduling
genetic algorithm
CloudSim platform
