Real-time Network Topology Discovery Techniques Of The Network Situation Awareness System

Nowadays, network is becoming more and more important in people's life with the development and popularization of the Internet. But network security situation are getting worse day after day. The present network security products can't aware network security situation accurately. Then the research on network situation awareness is carried on and is becoming hot in the field of network security. Network situation awareness is required to monitor real-time network situation. So network topology discovery in the network situation awareness system should have the ability of real-time discovery. Network topology discovery not only shows link structure of network but also provides a network situation display platform. Then manager can have an accurate and clear understanding of current network situation. Real-time network topology discovery is a research emphasis in this paper.In the paper, firstly status quo and main methods of network topology discovery are discussed. Secondly, after comparing various discovery methods, an algorithm of real-time network topology discovery based on SNMP protocol is raised and its several technologies are studied emphatically, including confirmation of network device's activity, distinguish of network device's type and storage and disposal of network topology information. According to the algorithm, real-time network topology discovery is designed and is implemented with Java. Finally, an experiment network is constructed to prove feasibility of the algorithm.The experiment proves that all the active network devices linked to a 2-layer switch can be detected and be accessed for relative information by SNMP protocol. Because data packets are sent to every network device to confirm their activity, multi-thread program is made to reduce detection time and improve performance of real time. Even so, some time will be spent and precision can't be counted by second. The experiment also proves that it will cost dozens of seconds to finish detecting more than ten network devices. The bigger number of devices is, the more time will be spent. Maybe precision will be counted by minute.