Measuring the variability of CAIDA internet traffic traces
Georgios Y. Lazarou, Mohammad Saad Alam, Joseph Picone · 2016
In this paper, we studied the variability of 17 CAIDA Internet traffic traces which were collected in 2013, 2014, 2015 and 2016. The variability of these traces was measured by using the Index of Variability. Based on the results, we outlined several important observations. In particular, the Index of Variability has the ability to reveal significant differences between traffic traces. It is dynamic and its behavior depends on several factors, such as network protocol dynamics and link speeds. In addition, traffic source link speeds have a major impact on network traffic variability (burstiness). Also, results show that there is a significant reduction in the variability for the 2015 and 2016 traces.