The Research On Virtual Machine Migration Based On Intelligent Algorithm In Cloud Computing

As a new type of model,cloud computing is gradually occupying markets in various fields.Cloud computing can simply be defined as a virtualized pool of computer resources,it can deploy,configure and recycle computer resources quickly to meet users' demands,and because of its flexibility,easy operation,low price and other characteristics,many companies have recognized it.However,with the expansion of data centers in cloud computing,the issue of high energy consumption and unbalanced allocation of resources have become a major obstacle to the development of cloud computing.It is imperative to reduce overall energy consumption and ensure the effective allocation of resources in cloud data centers by a viable strategy of VM migration.VM migration mainly has three choices: the source hosts,the virtual machines which should be migrated and the destination hosts.This dissertation mainly takes these three choices as a breakthrough point and makes use of related intelligent algorithms to make decisions.The main work is as follows:(1)Considering that a good initial placement is conducive to the subsequent operations of VM migration,this dissertation has proposed an energy-aware VM initial placement strategy based on BPSO.(2)Due to the fact that the demands for resources from virtual machines in a host is dynamically changing during the process of handling users' requirements in a cloud data center,this dissertation has proposed dynamic VM consolidation based On migrate consideration obtained by using Genetic Expression Programming.This strategy not only avoids the lack of local optimal solution caused by other strategies,but also can consider a variety of resources,which can increase the availability of the strategy.In addition,the algorithm introduced a new concept: migration consideration.It is used for computing the loss of energy during the whole process of VM migration.Therefore,the strategy selects the appropriate source hosts and destination hosts for VM migration which considers the loss of energy and the optimization of energy consumption.Finally,based on a cloud simulation platform of cloud computing: Cloudsim,this dissertation has implemented our strategies of VM initial placement and VM migration.Experimental results show that the proposed strategies have some success in reducing energy consumption.