Research And Implementation Of Energy - Based Virtual Resource Management In Cloud Environment

Since the concept of cloud computing being presented, it becomes the most exciting research focus rapidly in the field of science and technology. With the development of cloud computing, data center as the infrastructure and hardware of cloud computing, is also growing rapidly. Data center consumes huge amounts of energy, more and more people pay attention to the problem of the data center’s energy consumption. The research shows that the growth of data center power consumption is doubling every five years. While data center consumes large amounts of power, also produces CO2emissions and significantly contributes to the growing environmental issue of Global Warming. The problems of how data center to reduce energy consumption, reduce CO2emissions, has become an urgent problem need to be solved.In this thesis, from the point of servers’energy saving, we use the virtualization technology, virtual machine dynamic migration technology and cpu dynamic voltage/frequency scaling technology, to reduce the server’s energy consumption. In this thesis, we propose an energy-efficient virtual machine placement and adjustment mechanism to optimize the virtual machine, shutdown the idle physical machine, improve the utilization of physical resources, reduce server’s power consumption. In this thesis, we research the relationship between server’s cpu power consumption and cpu load, and propose a model of server’s power consumption. Through this model, we can approximate calculation of the power consumption. We also consider the virtual machine migration time and the loss of virtual machine while in the migration.In order to verify the effect of energy-efficient virtual machine placement and adjustment mechanism, we have designed and developed a cloud computing virtual resource management prototype system. This prototype system can create virtual machine via hypervisor, and monitor virtual machine and physical machine’s information. We add the energy-efficient virtual machine placement and adjustment mechanism into the prototype system, and finally verify the energy-saving effect through the experiment.