Lessons from "on the self-similar nature of ethernet traffic"

Abstract

This editorial is an outgrowth of our research efforts that resulted in the SIGCOMM'93 paper [1] entitled On the self-similar nature of Ethernet traffic. We discuss some lessons we have learned as we have watched the published findings being absorbed by the scientific community in general and the networking community in particular. We focus on aspects that have remained relevant today, especially at a time when, with the proliferation of Artificial Intelligence and Machine Learning, networking research has become increasingly data-dependent and data-driven.