Font Size: a A A

Research On Energy-aware Virtual Machine Migration Strategy In Cloud Data Center

Posted on:2020-12-03Degree:MasterType:Thesis
Country:ChinaCandidate:D SongFull Text:PDF
GTID:2428330590495410Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Cloud computing satisfies the needs of computing power from users by cloud data centers.With the bigger size of cloud data centers,the energy consumption of data centers becomes larger.In addition,the network traffic of communication between virtual machines has a large effect on the performance of servers in cloud data centers.Therefore,it is a research hotpot that reducing energy consumption and improving the quality of service by migrating virtual machines.To deal with the problem,the energy-aware virtual machine placement using multi-objective optimization is proposed in this thesis.When several data centers work together to provide service for users,it is not efficient to reduce energy consumption and ensure the quality of service merely by migrating virtual machines in a single data center.To achieve the target,the strategy of energy-aware virtual machine migration in multiple data centers is proposed.The main content of the thesis is energy-aware deployment of virtual machines using multiobjective optimization and the strategy of virtual machine migration in multiple data centers.The main work of the thesis is as follows: Firstly,to solve the problem of high energy consumption and network traffic,the energy-aware deployment of virtual machines using multi-objective optimization is proposed,the experiment results show that the algorithm performs better than traditional heuristic algorithms in virtual machine placement.Secondly,to deal with the problem of energy consumption in multiple data centers,a novel method is proposed: First,determine target data centers in virtual machine migrations across data centers according to the bandwidth between them.Then,determine overloading hosts and virtual machine selection policy according to the history CPU loads of data centers.As is shown in the experiment,the algorithm has a good performance in cutting down energy consumption and ensuring the quality of service.Finally,the module of virtual machine deployment on the OpenStack platform is implemented in this thesis.First,the OpenStack platform is built.Then,the module of virtual machine deployment on the OpenStack platform is designed and implemented by applying the virtual machine migration algorithm in a single data center in this thesis,proving the feasibility of the policy.
Keywords/Search Tags:Cloud Computing, Energy Consumption, Cloud Data Center, Load Balancing, Multi-objective Optimization
PDF Full Text Request
Related items