中文摘要：

针对云环境下虚拟机资源在多数时间中处于闲置状态导致云资源利用率低的问题,设计一种云资源监控系统,并在云监控基础上提出一种基于自回归积分滑动平均（ARIMA）模型的动态负载预测与资源配置的方法.该方法利用虚拟机负载与配置的关系,通过预测负载情况,提前启动或者挂起虚拟机,提高云资源的利用率.研究结合OpenStack云环境提供的虚拟机,实现其下的云资源监控,预测和弹性分配功能.结果表明：该系统能准确预测虚拟机的需求量,所制定的资源弹性分配策略能够提高云资源的利用率,进一步节约成本.

英文摘要：

In order to solve the problem under the cloud of virtual machine resources are idle most of the time,resulting in a cloud resource utilization is low,we designs a cloud resource monitoring system in this paper.And a dynamic load forecasting and resource allocation method is proposed based on cloud monitoring and a autoregressive integrated moving average（ARIMA）model.This method makes use of the relationship between the load and the virtual machine′s virtual hardware configuration by predicting loads and pre-starting or suspending a virtual machine to improve the utilization of cloud resources.Our studies realize monitoring,forecasting cloud resources and elastic cloud resources allocation based on the virtual machines in the OpenStack cloud environment.Finally,the result shows that：the system can accurately predict the demand for virtual machines and elastic cloud resources allocation policies can imporve the utilization of cloud resources and drive cost savings.