Cloud computing virtual machine migration method based on Frescher distance

Abstract

The invention discloses a cloud computing virtual machine migration method based on a Frescher distance. The method comprises the following steps of: screening out overload physical machines of a cloud data center, selecting a virtual machine to be migrated by adopting a least virtual machine migration algorithm, predicting resource use conditions of the virtual machine in the future t times, calculating residual resources of a host which is not overloaded, and predicting resource residual conditions of the host which is not overloaded in the future t times. And calculating the similarity between the resource use curve of the virtual machine to be migrated and the resource remaining curve of the overload physical machine by using the Frescher distance, and selecting one group of virtual machine and physical machine with the maximum similarity value to finish migration. According to the invention, the time distribution of virtual machine resource requests on a physical machine is analyzed. The resource use condition of the virtual machine in the future period of time and the resource remaining condition of the physical machine in the future period of time are predicted, so that theload balance of the migrated physical machine in the future period of time can be ensured. The migration frequency of the virtual machine in unit time is reduced, and the energy consumption of the cloud data center is reduced.

Description

technical field [0001] The invention belongs to the field of cloud computing and virtualization, and in particular relates to a cloud computing virtual machine migration method based on Fresche distance. Background technique [0002] Cloud computing technology is another sea change after the big shift from mainframe computers to client-server in the 1980s. It is the product of the development and integration of traditional computer and network technologies such as distributed computing, parallel computing, and network storage. Cloud computing adopts the mode of providing services on demand and paying according to usage. Under this mode, users can fully enjoy the flexibility of time and space brought by cloud computing. The flexibility of time is mainly reflected in the fact that cloud computing can be used at any time Provide services to users who need them; the flexibility of space is mainly reflected in the provision of resources according to the needs of users. [0003]...

AI Technical Summary

Problems solved by technology

[0005] In the cloud data center, physical machine overload often occurs, but the physical machine overload is not a continuous process, most of the overload is intermittent, because multiple virtual machines on the physical machine intensively request services during a specific period of time Most of the current migration strategies do not take into account the centralized request service of the virtual machine

The existing virtual machine migration strategy only migrates the physical machine when the physical machine is overloaded, but the resource usage of the virtual machine will change over time, which may cause the new physical machine to be overloaded shortly after the migration

This will increase the number of migrations and migration costs of the cloud data center

Images