摘要
针对当前OpenStack云平台的虚拟机调度策略未综合考虑服务器能耗、使用数量、计算性能带来的高能耗、服务器数量较多、较高的服务等级协议(SLA)违背率的影响,构建了综合这几方面因素的能效优化模型,根据该模型设计了一种基于自适应双策略差分进化算法的虚拟机放置策略。算法包括变异阶段根据种群分组差异选择不同变异策略,增加自适应交叉与变异参数用于提高算法收敛速度,使用综合能效优化模型作为适应度函数评价进化个体。实验仿真表明,本文策略使得OpenStack中虚拟机放置具有比其内置算法和DE、PSO算法更低的能耗、更小SLA违背率和更少的服务器使用数量。
Current scheduling strategies for virtual machines on OpenStack cloud platforms do not focus on high energy consumptio.Large number of servers,and high rate of service level agreement(SLA)violation were brought about by server energy consumption,and computing performance.This paper constructs an energy efficiency optimization model to integrate these three factors.Based on the model,a virtual machine placement strategy based on adaptive dual-strategy differential evolution algorithm is designed.This new strategy includes modifying differential evolution algorithm(DE)by citing different mutations way in different groups and self-adaption factors of mutation operation and cross operation.It can improve the convergence speed,and evaluate evolutionary individuals by integrated energy efficiency model as the algorithm fitness function.The simulation results show that the strategy has lower energy consumption and SLA violation rate,less number of servers used than built-in algorithms,DE algorithms,and PSO algorithms.
作者
罗平
王勇
俸晧
何倩
LUO Ping;WANG Yong;FENG Hao;HE Qian(School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China;Guangxi Colleges and Universities Key Laboratory of Cloud Computing and Complex Systems,Guilin University of Electronic Technology,Guilin 541004,China;School of Computer Science and Information Security,Guilin University of Electronic Technology,Guilin 541004,China)
出处
《桂林理工大学学报》
CAS
北大核心
2018年第3期555-560,共6页
Journal of Guilin University of Technology
基金
国家自然科学基金项目(61661015
61662018)
广西云计算与复杂系统高校重点实验室项目(2015209
14103
15208)
广西云计算与大数据协同创新中心研究课题(YD16303)
广西研究生教育创新计划项目(JGY2014062)
关键词
云计算
OPENSTACK
能耗
虚拟机放置
差分进化算法
cloud computing
OpenStack
energy consumption
virtual machine placement
differential evolution algorithm