An energy-efficient virtual machine migration method in a cloud data center

Abstract

The invention discloses an energy-efficient virtual machine migration method in a cloud data center. The method is cooperatively finished by four parts including physical machine overload monitoring, virtual machine selection, physical machine light load monitoring and virtual machine migration. The method comprises the following steps: firstly, collecting a use situation of physical machine resources, and utilizing a prediction technology to judge whether the physical machine is subjected to overload or not; for the physical machine subjected to overload, selecting the virtual machine which need to be migrated from the physical machine so as to lighten the load of the physical machine and improve the service quality of the data center; and after migration is finished, selecting the physical machine under a light load state from the data center, and shutting down the physical machine to achieve a purpose of energy saving after all virtual machines of the physical machine are migrated. The method can improve the service quality while the purpose of the energy saving of the data center is realized, and is suitable for the heterogeneous cloud data center.

Description

technical field [0001] The invention belongs to the technical field of the Internet, and in particular relates to an energy-efficient virtual machine migration method in a cloud data center. Background technique [0002] Cloud computing, developed from cluster, grid and utility computing, is a new paradigm in which current data centers provide computing resources dynamically. With the application of the payment model, a large number of computing resources and storage resources are placed in the cloud, and users can easily obtain and use these resources from the cloud according to their own needs and pay for them. This allows users to focus on the development of application programs, freeing them from low-level tasks such as configuring server hardware and managing system software, greatly improving work efficiency and reducing user development costs. [0003] With the continuous growth of the number of cloud users, a large number of cloud data centers have begun to emerge, ...

AI Technical Summary

Problems solved by technology

[0010] 3. For a large-scale cloud data center, the above UMC algorithm must traverse all physical hosts and calculate the correlation coefficient when selecting a host host for each virtual machine. The time cost of the algorithm is high, and the real-time performance cannot be guaranteed.

Images