Abstract
This paper criticises the notion that long-range dependence is an important contributor to the queuing behaviour of real Internet traffic. The idea is questioned in two different ways. Firstly, a class of models used to simulate Internet traffic is shown to have important theoretical flaws. It is shown that this behaviour is inconsistent with the behaviour of real traffic traces. Secondly, the notion that long-range correlations significantly affects the queuing performance of traffic is investigated by destroying those correlations in real traffic traces (by reordering). It is shown that the longer ranges of correlations are not important except in one case with an extremely high load.
Summary
Challenges the idea that long-range dependence is an important driver of real internet traffic queuing behaviour: shows theoretical flaws in a class of traffic models used to produce it, and demonstrates by reordering real traffic traces that destroying long-range correlations barely affects queuing performance except under extremely high load.
bibtex
@inproceedings{clegg2009criticismslrd,
author = {Richard G. Clegg and Raul Landa and Miguel Rio},
title = {Criticisms of modelling packet traffic using long-range dependence},
booktitle = {2009 Proceedings of the 18th International Conference on Computer Communications and Networks (ICCCN)},
year = {2009},
pages = {61--74},
doi = {10.1109/icccn.2009.5235309}
}
Authors
Richard G. Clegg, Raul Landa, Miguel Rio
Venue
2009 Proceedings of the 18th International Conference on Computer Communications and Networks (ICCCN), pp. 61-74