Internet Traffic tends to Poisson and Independent as the Load Increases
The burstiness of Internet traffic was established in pioneering work
in the early 1990s, which demonstrated that packet arrival times are
not Poisson, and packet and byte counts in fixed-length intervals
are long-range dependent. Our study of Internet traffic has shown that these
results are one end of a continuum of traffic characteristics. At the
other end are Poisson behavior and independence.
Packet Header Measurement   TCP/UDP/IP packet header collection, and organization of the headers into connection flows, is a powerful measurement framework for Internet traffic study. We have two major packet collection projects in place that have yield nearly 1 terrabyte of compressed headers that have been moved into our S-Net environment traffic data analysis.
S-Net   This traffic measurement and analysis environment, which begins with packet header collection on network wires, and ends with data analysis on a cluster of Linux PCs running S, is allowing analyses of very large Internet packet-header databases in great detail and breadth.
Fundamentals of Traffic Modeling   Our approach exploits the fundamental structure of Internet traffic --- the superposition of different traffic sources. Superposition underlies our study of the data and underlies our statistical models, which has led to a number of important discoveries about the fundamental nature of Internet traffic and to realistic, parsimonious models for traffic variables.
Statistical Modeling of Connection Request Variables   Our studies of connection request variables have led to a comprehensive modeling of these variables. The models are needed both for a fundamental understanding of Internet traffic and for providing realism in systems that generate packet-level Internet traffic.
PackMime   This packet-level Internet traffic generator mimics the behavior of packet traffic on live Internet wires. Our statistical models for connection variables generate requests for connections from a cloud of users; the requests are given to a network model that uses the source code of TCP from the BSD kernel to generate packets.
Helios Next-Generation-Internet Project   As part of this DARPA funded work our Helios team from Bell Labs Research, Lucent Optical Network Group, NCNI Research, and UNC Computer Science, is collecting and analyzing high-speed Internet traffic data on the NCNI gigapop. And we are implementing a S-Net-Helios environment at link speeds of up to 2.5 gigabits/second.
Visualization   Our studies of traffic depend heavily on visualization tools. We use a number of existing tools and have developed new ones. Because the traffic databases are large, single displays in our analyses can cover hundreds of pages. For a number of papers below, we provide the visualizations that we used in our analysis.
Papers about these and other Internet traffic research projects.