云计算环境下IDC虚拟化调度节能方案的设计与实现

发布时间：2018-05-03 19:38

本文选题：云计算 + 虚拟化　；参考：《复旦大学》2013年硕士论文

【摘要】：云计算的概念自从被提出以来就迅速成为计算机领域的研究热点,受到工业界和学术界的广泛关注。作为云计算的物理载体的数据中心(IDC)也随着云计算技术的不断发展而发展,在数据中心发展的同时,电能消耗的问题日益突出。研究表明,数据中心的电能消耗正以每5年翻一番的速度增长,电能消耗的同时会使数据中心排放大量的二氧化碳,对环境造成影响。如何让数据中心降低能耗,节能减排,已成为一个迫切需要解决的问题。本文从数据中心节能的角度出发,利用云计算中的虚拟化技术和虚拟机动态迁移技术,以及CPU的动态电压/频率调整(DVFS)技术,来降低数据中心的电能消耗。本文提出的云计算环境下IDC节能的虚拟化调度方案,通过优化虚拟机的放置方法,关闭空闲的物理机,来提高物理资源的利用率,降低数据中心的电能消耗。本文通过研究数据中心服务器的功耗与CPU负载之间的关系,提出了数据中心能耗估算模型,通过该模型可以利用CPU的负载来近似计算数据中心服务器的能耗。本文同时考虑了虚拟机迁移的时间以及迁移过程中造成的性能损失,提出了VM迁移代价模型以及SLA度量模型。本文所设计的虚拟化资源节能调度方案在某IDC的云计算环境中实施部署,该系统以物理服务器作为资源池,通过Hypervisor创建相互独立的虚拟机,细粒度的控制虚拟机的调度,同时监控数据中心中各个物理机和虚拟机的负载信息,通过未实施部署前的能耗数据与其他虚拟机调度算法的测试数据对比的结果,证明了本资源调度方案具有良好的节能效果。
[Abstract]:The concept of cloud computing has become a research hotspot in computer field since it was put forward, and has attracted wide attention from industry and academia. As the physical carrier of cloud computing data center IDC) is also developing with the development of cloud computing technology. With the development of data center the problem of energy consumption becomes more and more prominent. The research shows that the power consumption of the data center is doubling every five years, and the power consumption will make the data center emit a lot of carbon dioxide at the same time, which will have an impact on the environment. How to make the data center reduce energy consumption, energy saving and emission reduction has become an urgent problem to be solved. From the point of view of data center energy saving, this paper uses the virtualization technology in cloud computing, virtual machine dynamic migration technology and CPU dynamic voltage / frequency adjustment technology to reduce the power consumption of data center. In order to improve the utilization of physical resources and reduce the power consumption of data center, this paper proposes an energy-saving virtualization scheduling scheme for IDC in cloud computing environment. By optimizing the placement method of virtual machines and shutting down idle physical machines, we can improve the utilization of physical resources and reduce the power consumption of data centers. By studying the relationship between the power consumption of the data center server and the CPU load, a data center energy consumption estimation model is proposed in this paper. By using this model, the energy consumption of the data center server can be approximately calculated by using the load of the data center server. In this paper, the time of virtual machine migration and the performance loss caused by the migration are considered, and a VM migration cost model and a SLA metric model are proposed. The virtual resource energy saving scheduling scheme designed in this paper is deployed in a cloud computing environment of IDC. The system uses physical server as resource pool, creates independent virtual machine through Hypervisor, and controls the scheduling of virtual machine fine-grained. At the same time, the load information of each physical machine and virtual machine in the data center is monitored and compared with the test data of other virtual machine scheduling algorithms. It is proved that the resource scheduling scheme has good energy saving effect.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP308;TM92

【相似文献】