Based On The NS2Large-scale Network Service Simulation And Analysis

As the network continues to increase, the scale of the network service categoryanalysis become more and more complex. And autonomous computing dealing withincreasingly complex network environment, security software system status servicequality become the most important method.The autonomic computing is an important method to deal with increasingly complexnetwork environment, ensure the quality of software system. Machine learning is aneffective way of achieving functional autonomy of software system. Quantitativestructural data has significant implication for machine learning, but in the actual systemthe acquisition of quantitative structural data is difficult, affecting the research of machinelearning methods for functional autonomy. In this paper, a simulation method ofquantitative structural data is introduced which based on NS2network simulation platform,this method can simulate service state in different network topologies under different typesof service requests, and then extract and generate quantitative structural data diagram, thediagram serves the machine learning method research which implements the functionalautonomy. Aiming at the large scale network topology structure is proposed based on theadjacency model storage of a hierarchical recursive network status hierarchy analysismethod, in order to facilitate under imperfect information reasoning in decision making,problem solving and research. In this article the application background of large-scalenetwork, extensive simulation data set and network quantitative structure level iterativecharacteristics, through the correlation method and the known subsystem rule sets, to builda massive hierarchy analysis method, finally the paper gives the model of basic instanceand analytic hierarchy process method process.Experimental results show that the network simulation method has good operationand simulation results, and can effectively carry out the scale and level of expansion,providing the basis for large-scale functional autonomy analysis of the network. In thisarticle, the application of underlying network simulation data with the large-scale networkservice status analysis, in this way, it can be show the state of high level network node andgive a better state result. At the same time, in this article further through regular scanning of network status, network status which was exception can recovery itself, and then markthe network status flag normal, so that the status of network services can be guaranteed tobe in the normal state which running in a fairly long period of time.