中文摘要：

针对云计算服务环境下静态阈值的虚拟机（Virtual Machine,VM）动态迁移算法无法根据云任务负载实时优化系统能耗和负载均衡的问题,（方法）提出利用工作负载历史数据的统计学规律自适应地确定迁移阈值,实现云计算环境下虚拟机（Virtual Machine,VM）动态调度的自适应节能算法.（结果）采用动态电压频率调节（dy-namic voltage frequency scaling,DVFS）--实现系统部件静态节能,又通过两种自适应主机超载判定（Adaptive Host Overloading Detection,AHOD）算法实现云数据中心的 VM 动态迁移,在 Cloudsim 云仿真平台下对比实现 DVFS静态节能和2种 AHOD 动态节能策略,（结论）结果表明：DVFS 和 AHOD 策略可以显著节能68%以上,虽然AHOD 策略中 VM 动态迁移引起的服务等级协定违例（Service Level Agreement Violation,SLAV）将导致云服务质量（Quality of Service,QoS）降低,并造成数据中心综合性能下降,然而合理建模并找到能耗与 QoS 的平衡点,即安全阈值为1.2的 AHOD 算法不但能够实现自适应负载均衡的虚拟机动态迁移,还实现云环境下的高效节能.

英文摘要：

In cloud computing environment, Virtual Machine （VM） dynamic migration algorithm with static threshold could not optimize the energy consumption and load balancing with real-time cloud workload.In response to these problems,the proposed work defined the migration threshold adaptively using the statistical regularity of workload historical data, in order to achieve the adaptive energy-efficient VM scheduling algorithm in cloud computing environment.Dynamic voltage frequency scaling （DVFS） was adopted as the static energy saving strategy,while VM live migrations with 2 AHOD （Adaptive Host Overloading Detection）algorithms were adopted as the dynamic energy saving strategy in cloud data center.After the simulation and contrast to the static energy saving of DVFS and the dynamic energy saving with AHOD algorithms,the results show that：DVFS and AHOD strategy can significantly save energy 68% or more,although VM dynamic migration of AHOD strategy caused SLAV （Service the Level Agreement Violation）,which lead to the degradation of QoS （Quality of Service）and the overall performance of the data center in cloud computing.However reasonable modeling is able to balance the energy consumption and QoS,security threshold of 1.2 in AHOD algorithm could consume the energy efficiently, achieve the adaptive load balancing of virtual machines migration dynamically.