Performance trending for management of high-speed packet-switched networks

Due to technological advances, computer communication networks are becoming increasingly complex. In addition, applications require networks to provide high performance and reliability. Placing such demands on these networks necessitates an efficient network management. Many existing systems perform data collection and reporting only, and so operators must use their own expertise to do aid the trouble shooting. To provide a dynamic solution, it is necessary to provide a real-time automated control of network operation parameters. The objective of this research is to develop and demonstrate the capability of applying modeling and simulation techniques to maintain the network integrity of a high-speed packet-switched network. In the paper a new technique called performance trending is presented, in which key network parameters are monitored and network failure is predicted. Performance trending is also used for the network performance maintenance by means of analysis and tuning to predict and prevent network failures. A typical packet-switched network is simulated to verify the performance trending for network performance management. End-to-end delay and bit error rate (BER) are used to illustrate performance trending and their effects on the network performance