| Summary: | 碩士 === 中華大學 === 資訊工程學系 === 105 === Cloud computing has been the global concern of the computing environment, and cloud computing is a large-scale distributed computing. Virtualization technology allows entities to use resources more efficiently. With the increasing popularity of container virtualization technology,and Container as a Service (CaaS) has been widely used. As the demand for cloud computing increases, the amount of information that causes cloud computing increases. Data center energy consumption is undoubtedly becoming larger for cloud providers to increase costs. One of the major challenges that cloud providers face is minimizing power consumption of their data centers. In this paper, we consider energy consumption for reduce the cost of data centers, so that the data center to effectively reduce the power consumption to achieve the purpose of energy conservation. The study shows that the usage rate of the physical machine is non-linear with the energy consumption. We use precise mathematical models and analysis,and configure the work to fit the machine according to the size of the memory. We propose three methods for performance analysis. The experimental results show that the appropriate method can be selected according to different types of work, and can effectively reduce energy consumption. This study can be widely used in the data center, for the development of CaaS also have some reference value.
|
|---|
