Analysis of Network Traffic Based on Flow Level
Zhongcheng Li · Mini-micro Systems · 2005
The analyzing characteristic of network traffic is the base to prompt network performance. The self-similarity is a common characteristic of network traffic. Through the analyzing characteristics of backbone network traffic at flow level, it shows that the self-similarity of the time series of flow intervals at small time scales is weaker whereas it is stronger at large time scales. The further study shows the size of flows and the ICMP flows apparently affects the self-similarity. Especially, those flows of one packet have larger effects.