EXPERIMENTAL QUEUING ANALYSIS WITH LONG-RANGE DEPENDENT PACKET TRAFFIC

Recent traffic measurement studies from a wide range of working packet networks have convincingly established the presence of significant st atistical features that are characteristic of fractal traffic processe s, in the sense that these features span many time scales, Of particul ar interest in packet traffic modeling is a property called long-range dependence (LRD), which is marked by the presence of correlations tha t can extend over many time scales, In this paper, we demonstrate empi rically that, beyond its statistical significance in traffic measureme nts, long-range dependence has considerable impact on queueing perform ance, and is a dominant characteristic for a number of packet traffic engineering problems, In addition, we give conditions under which the use of compact and simple traffic models that incorporate long-range d ependence in a parsimonious manner (e.g., fractional Brownian motion) is justified and can lead to new insights into the traffic management of high speed networks.