Research And Implementation Of Virtual Machine Management And Deployment Technology For Ocean Cloud Platform

Cloud computing technology has caused a revolution in the information technology industry,which has become a hot research field in the industry and academia.Virtualization technology has become the core technology of cloud computing,because it has many advantages such as enhancing the flexibility and scalability of the system,and improving the utilization of resources.While the virtual machine management and deployment technology is the key technology in the field of virtualization.It has become an important research topic in virtual machine management and deployment that how to use the appropriate strategy and algorithm to deploy the virtual machine quickly to the cluster of the cloud platform,and to ensure that the virtual machine runs stably on the cluster.The existing deployment strategies and algorithms have less consideration during the process of virtual machine deployment,which can easily lead to unbalanced load and low resource utilization.That is difficult to meet the actual needs of the ocean cloud computing.In order to solve the above problems,combined with the actual needs of the National Marine Cloud Computing project,this thesis deeply research on the virtual machine resource management and deployment technology.Themain contributions can be summarized as follows.(1)In order to solve problems of load balancing and low resource utilization in the existing virtual machine deployment algorithms,this thesis put forward a virtual machine deployment algorithm based on characteristic matching,based on the deep research of the existing virtual machine deployment algorithms.Taking into account the memory,CPU and other factors of the cluster,the algorithm are the comprehensive resource characters matching results with the cosine similarity combined and Euclidean distance,and the virtual machine multi-objective optimization is realized.The algorithm realized multi-objective deployment of virtual machine.Compared with the traditional deployment algorithms,this algorithm achieves better load balancing effect as well as higher resource utilization and better energy saving effect.(2)By using the virtual machine self-service management architecture of the virtual machine deployment algorithm proposed in this thesis,the automation of the virtual machine deployment process can achieve the overall load balance,energy saving effect and so on.Through deeply research of open source virtualization management tools,we develop the virtual machines self-service application system.With application of the system,users can manage their own virtual machine based on Web,such as landing,closing,delete,etc.This can reduce the burden on administrator,and improve the operation and maintenance efficiency of the system.(3)The availability of the proposed algorithm is tested and compared with the first fit algorithm based on the practical application.The experimental results show that the proposed virtual machine deployment algorithm can achieve better load balancing effect and resource utilization,while the better energy-saving effect is obtained.In addition,The function test of the virtual machine self-service application system is carried out,including virtual machine application,list of virtual machines,and virtual machine boot start and other related management functions.Results show that the system achieves the design goals and requirements.