基于能耗优化的云虚拟机动态放置方法.docVIP

基于能耗优化的云虚拟机动态放置方法 苏楠 （武汉纺织大学 数计学院） 摘 要　针对如何将虚拟机合理的放置在服务器上，实现云环境下资源的高效分配并兼顾减低能耗这一问题，提出了一种基于模拟退火算法的动态云资源分配方法。首先，研究和建立了服务器运行能耗模型和虚拟机动态放置模型，为能耗优化研究提供基础。然后综合考虑系统性能要求、能源消耗等因素，将资源分配与能耗优化结合起来，形成虚拟资源分配能耗优化模型。最后使用模拟退火算法，自适应地计算出分配权重向量，有效地提高资源利用率，从而降低能耗。对比实验结果表明，该方法能在较短的时间内确定虚拟机放置策略，给定虚拟资源相同情况下，完成用户任务的同时能耗更低。 关键词 动态放置；能耗优化；模拟退火；资源利用率 中图法分类号　TP312　　 文献标识码　A An Consumption of Energy Optimization Based Virtual Machine Vitual Placement Method in the Cloud Computing Abstract　To address the issue of how to place the virtual machine on an appropriate server in the circumstances of cloud computing to achieve high allocation efficiency and low energy consumption, a dynamic cloud resource allocation method is proposed in this paper which is based on simulated annealing algorithm. First, build a server running virtual machine energy consumption model and a dynamic placement model for the study of the optimization of energy consumption. Then by considering the system performance requirements, energy consumption and other factors, the allocation of resources and energy optimization are combined to form a virtual resource allocation optimal model of energy consumption. Finally， simulated annealing algorithm is used to adaptively determine the calculation weight vector for virtual machine allocation in the aim of resource utilization improvement and energy consumption reduction. Comparative experimental results show that the proposed method can determine virtual machine placement strategy in a short time, furthermore，it can complete user tasks with lower energy cost given the same circumstances of virtual resources. Keywords　Dynamic allocation ; Optimization of Energy consumption; Simulated Annealing algorithm; Utilization of resource 0 引言 随着计算机和互联网技术的高速发展，服务器承载的信息量和访问量成几何级数增长[ ]，云计算作为一种新的模式应运而生。云数据中心的规模也逐步增大，高能耗成为云数据中心资源管理的一大挑战，这是因为在大规模云环境中，能耗是主要的运营成本因素。除了服务器运行的正常能耗以外，散热需要耗费额外的电能，高能耗不但过度消耗电能，且容易造成局部过热，影响整个系统的稳定性。而系统资源分配的不合理是导致能耗过高的一个主要原因[ ]，因此高