DRAFT On the Self - Similar Nature of Ethernet Traffic

We demonstrate that Ethernet local area network (LAN) traffic is statistically self-similar, that none of the commonly used traffic models is able to capture this fractal behavior, that such behavior has serious implications for the design, control, and analysis of high-speed, cell-based (B-ISDN) networks, and that aggregating streams of such traffic typically intensifies the selfsimilarity ("burstiness") instead of smoothing it. Intuitively, the critical characteristic of this self-similar traffic is that there is no natural length of a "burst": at every time scale ranging from a few milliseconds to minutes and hours, similar-looking traffic bursts are evident. Our conclusions are supported by a rigorous statistical analysis of hundreds of millions of high quality Ethernet traffic measurements collected between 1989 and 1992, coupled with a discussion of the underlying mathematical and statistical properties of self-similarity and their relationship with actual network behavior. We also consider some implications for congestion control in B-ISDN and present traffic models based on self-similar stochastic processes. Selfsimilar traffic models provide simple, accurate, and realistic descriptions of traffic scenarios encountered during B-ISDN deployment.