Research And Application Of Algorithm And Disaster Recovery Strategy In Cluster System Based On Virtualization

The rapid development of computer technology and network technology has spawned a series of information technologies that change our lives,including virtualization technology and cluster technology.As a combination of virtualization technology and cluster technology,cluster system based on virtualization has so many advantages,such as simplifying infrastructure,realizing server integration,and reducing management costs.And it has become widely researched and applied in various computer related fields.However,many of these virtualization-based cluster systems still have problems such as low utilization rate of resources and insufficient security.Therefore,this paper studies the algorithm and disaster recovery strategy in the cluster system based on virtualization,and attempts to propose improvement scheme which not only improve the utilization rate of resources but also ensure security.Firstly,this paper studies the present situation of virtualization-based cluster system research all over the world,and summarizes the work and research results of predecessors.Next,this paper studies and understands the key technologies and related knowledge needed in this paper,including virtualization technology and cluster technology.Taking into account the specific environment of the laboratory,this paper chooses the cyberspace security training platform to be the research object.Then,according to the specific characteristics of the cyberspace security training platform,this paper designs and implements improvement scheme of the virtual machine allocation load balancing algorithm,the disaster recovery strategy when a single server is unexpectedly down,and the multi-cluster system server sharing.Finally,this paper deploys the software and hardware environment of the virtualization-based cluster system on the cyberspace security training platform in the laboratory and verifies the feasibility and effectiveness of the improvement scheme by experimental tests.The main work in this paper is as follows:(1)Read a large amount of literature,study the resear-ch results of cluster system based on virtualization and related fields all over the world,and summarize some of predecessors' work.(2)Study the principles,development,and application of the key technologies and related knowledge needed in this paper,including virtualization technology and cluster technology.(3)Optimize and improve the virtual machine allocation load balancing algorithm of the cluster system based on virtualization for the problem of low resource utilization rate.(4)Optimize and improve the disaster recovery strategy when a single server is unexpectedly down of the cluster system based on virtualization for the problem of insufficient security.(5)Optimize and improve the strategies for the simultaneous operation of multiple different cluster system based on virtualization for the problem of idle and shortage of scientific research equipment in computer rooms and laboratories in universities.(6)Design and implement multiple sets of experiments for the proposed improvement scheme,and analyze the experimental test results in depth to verify the effectiveness and feasibility of the improvement scheme.