Research On Key Technology Of Distributed Scanning For Networking Oriented Equipment

With the popularity of the Internet technology,people access to the Internet become more and more convenient.The Network equipment in people’s lives can be seen everywhere.However,the frequent occurrence of user information leakage in society has threatened the people’s lives and property safety.In order to discover the system vulnerabilities in the network in advance,do network protection,the role of distributed network scanning technology has become increasingly prominent.In this paper,the load balancing algorithm and the commonly used dynamic load balancing algorithm Pick-Kx are studied.The algorithm has good versatility.However,there are some shortcomings in the practical application.The algorithm only considers the machine load to calculate the task distribution ratio,without taking into account the computational power,the impact of task load.In order to adapt to the distributed scanning better,this paper improves and optimizes the existing problems of the algorithm,and proposes an improved load balancing algorithm called DTPick-Kx.The DTPick-Kx algorithm improves the server task distribution ratio calculation method,taking full account of the current distributed node calculate ability,machine load,task load.At the same time,the Pick-Kx algorithm does not solve the problem of load imbalance caused by task gathering.The DTPick-Kx algorithm designed task migration strategy to solve the problem.This strategy effectively addresses the load imbalance problem by migrating the task of the overloaded node to the lighter node.The experimental results show that the DTPick-Kx algorithm has better performance than the Pick-Kx load balancing algorithm in the average machine load and average response time.Aiming at the difference of target connectivity between different network nodes,the accuracy of scanning results is affected.This paper presents an efficient scanning node selection model based on IP address database update strategy.This evaluation model provides a way to select efficient nodes in a distributed network scan.At the same time,this model proposes an update strategy based on IP address database for the selection of update area.The strategy updates the probability by scanning the task scheduling times and intervals,selects the updating region according to the probability,and combines the exponential smoothing algorithm to fuse the data.The data is acquired by timing detection of the target area.Theexperimental results show that the proposed model can reduce the scanning time and improve the scanning accuracy.