The Risk Assessment Of Network Node Resources Based On The Joint Distribution Function

In recent years,the rapid development of the network has provided great convenience for people.Because of the fragility of the computer system and the openness of the network,the problem of network security is becoming more and more obvious.Excessive network attack and network behavior will bring network resource risk.The instability of network traffic often hinder the use of normal demand,occupy the resources of other activities,destroy the hardware devices such as CPU,memory and hard disk of computer servers,and bring risks and hidden dangers for normal use.Therefore,the security of the network needs more perfect monitoring and management measures to ensure the resources of network nodes running in a reasonable and stable range.The risk assessment of network node resources can predict the size of network risk in advance,and it can also analyze and assess the risks of network nodes when there is no risk.Therefore,the risk assessment of network node resources is important research significance.This topic has been studied on the risk assessment of network node resources.Through analyzing the research status of network security risk assessment at home and abroad,we know that most of the past network security risk assessment is aimed at single index,while ignoring the correlation between indicators.In view of these shortcomings,this paper selects multiple indicators of network node resources as the research object,and uses joint distribution function to describe the correlation between indicators.The risk assessment of network nodes based on joint distribution function of resource variables is proposed.The research is mainly based on three dimensional Copula function and Monte Carlo method to describe the correlation between each index of network node resources,and evaluate their network risks.In this paper,network node resources risk assessment of three dimensional Copula function based on joint distribution function is firstly studied.In order to reflect the inherent relevance of network node resources,we need to construct their joint distribution function when the risk assessment of network node resources is carried out.In this paper,a method of constructing joint distribution function is studied.From the point of network node resources,we use memory utilization rate,network traffic and CPU utilization as the research indicators,and construct the joint distribution function of them by using the three-dimensional Copula function.Compared with the traditional method,the Copula function can describe the relationship between network node resource indicators,get the precise value of joint distribution function between the indicators.We also introduce risk return period of the introduction of cyber source network node resources to assess the risk of network node resources.Thus providing a new risk warning mechanism for risk warning network node resource.In addition,when constructing the joint distribution function of network node resources,we can choose a more rapid and concise Monte Carlo method to construct the joint distribution function of network node resources.When constructing joint distribution function for higher dimensional data,the computation of Copula function will become more and more complex,and the selection of Copula function will become more and more complicated.Monte Carlo's method provides another way to construct joint distribution function.It does not need to show the joint distribution function of network node resources by using the display formula.It is simple and fast,and it is easy to get numerical results.At last,we introduce the resource risk function to evaluate the risk of network node resources.The experimental results show that the 3D Copula function and Monte Carlo method can simple and rapidly conduct the joint distribution function of network node resource.The resource risk function and network risk return period of network resources also can correctly evaluate the network node resources risk.The assessment results can more accurately reflect the change of network security situation,which provide a new study on the risk assessment method for network node resources.