Title :
A network traffic model based on measurement
Author :
Zhang, Dong-Yan ; Hu, Ming-Zeng ; Zhang, Hong-Li ; Kang, Ting-Biao
Author_Institution :
Res. Center of Comput. Network & Inf. Security Technol., Harbin Inst. of Technol., China
Abstract :
Network traffic and the value of performance measurement metrics can be forecasted using the traffic model. In this paper, after investigating model ARIMA which is often used in time series scopes and random process, a specific and effective method for performance forecast is presented. Through the measurement on the LAN, the data of experiment shows that the traffic model has the advantage of more forecasting precision and the traffic model is very suitable to practical operational networks.
Keywords :
autoregressive moving average processes; local area networks; performance evaluation; telecommunication traffic; ARIMA model; LAN; network traffic forecasting; network traffic model; performance forecast; performance measurement metrics; random process; time series; Computer networks; Economic forecasting; Electronic mail; Information security; Performance analysis; Predictive models; Random processes; Telecommunication traffic; Traffic control; White noise; Traffic model; forecasting precision; network measurement; traffic forecasting;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527612
