On the self-similar nature of Ethernet traffic (extended version)
IEEE/ACM Transactions on Networking (TON)
Wide area traffic: the failure of Poisson modeling
IEEE/ACM Transactions on Networking (TON)
Experimental queueing analysis with long-range dependent packet traffic
IEEE/ACM Transactions on Networking (TON)
IEEE/ACM Transactions on Networking (TON)
Self-similarity in World Wide Web traffic: evidence and possible causes
IEEE/ACM Transactions on Networking (TON)
Self-similarity and heavy tails: structural modeling of network traffic
A practical guide to heavy tails
Self-Similar Network Traffic and Performance Evaluation
Self-Similar Network Traffic and Performance Evaluation
On the relationship between file sizes, transport protocols, and self-similar network traffic
ICNP '96 Proceedings of the 1996 International Conference on Network Protocols (ICNP '96)
End-to-end microflow performance measurement of IPv6 traffic over diverse wireless topologies
WICON '06 Proceedings of the 2nd annual international workshop on Wireless internet
Lognormal and Pareto distributions in the Internet
Computer Communications
Wavelet analysis of long-range-dependent traffic
IEEE Transactions on Information Theory
Understanding Internet traffic streams: dragonflies and tortoises
IEEE Communications Magazine
Wireless Personal Communications: An International Journal
| Hi-index | 0.00 |
It is by now widely accepted that the arrival process of aggregate network traffic exhibits self-similar characteristics which result in the preservation of traffic burstiness (high variability) over a wide range of timescales. This behaviour has been structurally linked to the presence of heavy-tailed, infinite variance phenomena at the level of individual network connections, file sizes, transfer durations, and packet inter-arrival times. In this paper, we have examined the presence of fractal and heavy-tailed behaviour in a number of performance aspects of individual IPv6 microflows as routed over wireless local and wide area network topologies. Our analysis sheds light on several questions regarding flow-level traffic behaviour: whether burstiness preservation is mainly observed at traffic aggregates or is it also evident at individual microflows; whether it is influenced by the end-to-end transport control mechanisms as well as by the network-level traffic multiplexing; whether high variability is independent from diverse link-level technologies, and whether burstiness is preserved in end-to-end performance metrics such as packet delay as well as in the traffic arrival process. Our findings suggest that traffic and packet delay exhibit closely-related Long-Range Dependence (LRD) at the level of individual microflows, with marginal to moderate intensity. Bulk TCP data and UDP flows produce higher Hurst exponent estimates than the acknowledgment flows that consist of minimum-sized packets. Wireless access technologies seem to also influence LRD intensity. At the same time, the distributions of intraflow packet inter-arrival times do not exhibit infinite variance characteristics.