Design And Implementation Of Campus Network Monitoring And Alarming System

With the expansion of the campus network scale and the continuous development of the network applications, the network infrastructure has become increasingly complex. It's urgent for real-time monitoring, raising alarm and analyzing comprehensively. Network monitoring and alarming is an effective means to understand the status of the whole network. It can detect network congestion and other network anomalies and also achieve fault location based on detailed monitoring information. Network managers can fix the problem timely.Using network management theory and practical methods, this paper designed and implemented the campus network monitoring and alarming system based on Cacti in order to enhance network performance, optimize for configuration and improve service quality. The main contributions of this paper are as follows:1. Analyzed the demands of Donghua University campus network monitoring and alarming systems, analyzed the structure of Donghua University campus network. Based on that, this paper designed and implemented a campus network monitoring and alarming system. Monitoring and alarming system is divided into two parts. Monitoring part includes:getting data through SNMP protocol and storing the data collected in the form of RRD, getting network equipment logs through SYSLOG protocol and storing logs in the server, showing the data collected to the users through Web. Alarming part includes:up/down time status alarming, threshold alarming, log alarming, etc.2. Analyzed the network traffic load and congestion. The campus network optimization recommendations were given based on CPU utilization aggregation data and traffic aggregation data.3. Analyzed the real collected data of campus network based on Holt-winters exponential smoothing method. Used the RRDTool extension tools to apply Holt-winters method to the real network traffic prediction. Holt-winters method is proved to be better than the threshold prediction method.4. Proposed a method to analyze campus network traffic based on RP and RQA. Provided a new modeling method for campus network traffic data.The data show that the design of the system is reliable and with good performance. With the co-operation of monitoring part and alarming part, this system achieved the mission of campus network monitoring and alarming. Through analysis, network optimization recommendations are given to improve the campus network conditions. This paper used the Holt-winters exponential smoothing methods and RQA method based on Chaos theory to analyze traffic data, and achieved good experimental and actual results. The traffic data.will be predicted with more intelligence.