Research On Virtual Machine Security Technology Of Internet Of Things PaaS Platform In Oil And Gas Production

In this paper,based on the existing environment of Petro China Jilin oilfield company,Cloud Foundry open source components are used to build a PaaS platform for the Io T of Oil and Gas Production.Various applications of the platform are deployed in multiple Centos virtual machines.During running the process,it is found that the kernel of the virtual machine is very vulnerable to attacked by rootkit malicious programs.In response to this problem,this article has designed a virtual machine monitoring system which can monitor whether the virtual machine in the platform is in a safe state.When a kernel rootkit attacks an ordinary virtual machine in the platform,the virtual machine monitoring system can immediately issue a warning to remind the administrator,at the same time,the monitoring system can take out some real-time indicator information to build an indicator set,and the administrator can find malicious programs and process them in time,the monitoring system also contains an adaptive algorithm to optimize the monitoring model.Since the virtual machine monitoring system can only monitor the ordinary virtual machines in the platform in real time,and cannot predict and evaluate the virtual machine’s security status in the future,therefore,in this paper,the index information of the virtual machine monitoring system is assembled into an index set,and the BP neural network algorithm is used to predict whether the ordinary virtual machines in the platform will be safe in the future.In order to solve the shortcomings of BP neural network convergence slow and easy to fall into local optimum,this paper proposes to use particle swarm optimization to optimize BP neural network.However,the traditional PSO algorithm has a weak global search capability,and it is easy to fall into "precocity".Therefore,this paper improves the traditional particle swarm algorithm,improves the optimization ability of the particle swarm algorithm,and prevents the group from falling into "premature convergence".Then use the improved PSO algorithm and BP neural network to effectively predict the future state of the virtual machine in the platform,and complete the virtual machine state prediction model of the PaaS platform.