The Design And Implementation Of The Security Emergency And Rapid Recovery Management System Of The Website Under Cloud Environment

With the rapid development of cloud computing and virtualization technique, the website cluster application systems based on cloud computing have become more and more large-scaled and complex. Usually, large business will be deployed on enormous amount of virtual machines. Therefore, the security of storage operation center and top business application is extremely important.Firstly, the thesis investigates the balance strategy of Swift virtual nodes with the assumption that virtual nodes are distributed uniformly. However, the capacity of virtual nodes mapping object and the difference between access loads can not be ignored, so based on the investigation, we propose an improved strategy. It is a multi-objective dynamic load balancing of storage nodes strategies according to the capacity and access load of virtual nodes mapping objects. The experiment results show that the improved Ring can make the distribution of Swift cloud storage resources more uniform distributed, therefore, the proposed strategy can overcome the imbalance problem.Further, based on the IaaS cloud platform, cloud storage center deployment architecture and SDN network virtualization framework, combined with virtualization technology and web security technology, this thesis proposes a new security model for emergency response and rapid recovery strategy. The model aims to achieve security emergency and rapid recovery of large-scaled website cluster under cloud environment. The implementation is achieved through SDN intelligent traffic guidance, network migration and file integrity protection technology. It can make an emergency response and rapid recovery to a security threat business machine, and can provide efficient, safe and stable operating environment for the website cluster and realize large-scale high availability web application server.Moreover, based on the rapid recovery strategy model, a high availability strategy model which based on database of dynamic data is put forward. In this model, a proxy database server keeps sending date requests to multiple databases server with the same image of the original database server. And after all the data back to the original database proxy database server, the model uses data arbitration algorithm for data verification, to ensure high availability, accuracy and security for dynamic database in real-time transmission.Finally, requirement analysis of safety emergency and rapid recovery management system is made in this thesis. Besides, the key module safety emergency and rapid recovery subsystem have been designed and implemented. The subsystem includes OpenStack cloud platform, Swift cloud storage centers, Floodlight controllers and rapid response recovery strategy and other functional modules. Eventually, the performance of the subsystem is tested to meet the demands of large-scaled applications.