Contradictory relationship between Hurst parameter and queueing performance (extended version)

Long Range Dependent (LRD) network tra c does not behave like the tra c generated by the Poisson model or other Markovian models. From the network performance point of view, the main di erence is that LRD tra c increases queueing delays due to its burstiness over many time scales. LRD behavior has been observed in di erent types and sizes of networks, for di erent applications (eg. WWW) and di erent tra c aggregations. Since LRD behaviour is not rare nor isolated, accurate characterization of LRD tra c is very important in order to predict performance and to allocate network resources. The Hurst parameter is commonly used to quantify the degree of LRD and the burstiness of the tra c. In this paper we investigate the validity and e ectiveness of the Hurst parameter. To this end, we analyze the UCLA Computer Science Department network tra c traces and compute their Hurst parameters. Queueing simulation is used to study the impact of LRD and to determine if the Hurst parameter accurately describes such LRD. Our results show that the Hurst parameter is not by itself an accurate predictor of the queueing performance for a given LRD tra c trace.