History of network traffic models

Design of robust and reliable networks and network services relies on an understanding of the traffic characteristics of the network. Throughout history, different models of network traffic have been developed and used for evaluating existing and proposed networks and services.

Demands on computer networks are not entirely predictable. Performance modeling is necessary for deciding the quality of service (QoS) level. Performance models in turn, require accurate traffic models that have the ability to capture the statistical characteristics of the actual traffic on the network. Many traffic models have been developed based on traffic measurement data. If the underlying traffic models do not efficiently capture the characteristics of the actual traffic, the result may be the under-estimation or over-estimation of the performance of the network. This impairs the design of the network. Traffic models are hence, a core component of any performance evaluation of networks and they need to be very accurate.

“Teletraffic theory is the application of mathematics to the measurement, modeling, and control of traffic in telecommunications networks.[1] The aim of traffic modeling is to find stochastic processes to represent the behavior of traffic. Working at the Copenhagen Telephone Company in the 1910s, A. K. Erlang famously characterized telephone traffic at the call level by certain probability distributions for arrivals of new calls and their holding times. Erlang applied the traffic models to estimate the telephone switch capacity needed to achieve a given call blocking probability. The Erlang blocking formulas had tremendous practical interest for public carriers because telephone facilities (switching and transmission) involved considerable investments. Over several decades, Erlang’s work stimulated the use of queuing theory, and applied probability in general, to engineer the public switched telephone network. Teletraffic theory for packet networks has seen considerable progress in recent decades.[2][3][4][5] Significant advances have been made in long-range dependence, wavelet, and multifractal approaches. At the same time, traffic modeling continues to be challenged by evolving network technologies and new multimedia applications. For example, wireless technologies allow greater mobility of users. Mobility must be an additional consideration for modeling traffic in wireless networks.[6][7] Traffic modeling is an ongoing process without a real end. Traffic models represent our best current understanding of traffic behavior, but our understanding will change and grow over time.”[8]

  1. ^ Willinger and Paxson (1998). "Where Mathematics Meets the Internet" (PDF). AMS.
  2. ^ Park, Kihong; Willinger, Walter (2000). Self-similar network traffic and performance evaluation. New York: Wiley. doi:10.1002/047120644X.fmatter_indsub. ISBN 978-0-471-31974-0.
  3. ^ Adas, A. (1997). "Traffic models in broadband networks". IEEE Communications Magazine. 35 (7): 82–89. CiteSeerX 10.1.1.23.1461. doi:10.1109/35.601746. ISSN 0163-6804.
  4. ^ Michiel, H.; Laevens, K. (1997). "Teletraffic engineering in a broad-band era". Proceedings of the IEEE. 85 (12): 2007–2033. doi:10.1109/5.650182. ISSN 0018-9219.
  5. ^ Frost, V.S.; Melamed, B. (1994). "Traffic modeling for telecommunications networks". IEEE Communications Magazine. 32 (3): 70–81. doi:10.1109/35.267444. ISSN 0163-6804. S2CID 19019323.
  6. ^ Chien-Hsing Wu; Huang-Pao Lin; Leu-Shing Lan (2002). "A new analytic framework for dynamic mobility management of PCS networks". IEEE Transactions on Mobile Computing. 99 (3): 208–220. doi:10.1109/TMC.2002.1081756.
  7. ^ Thajchayapong, S.; Peha, J.M. (2006). "Mobility patterns in microcellular wireless networks" (PDF). IEEE Transactions on Mobile Computing. 5 (1): 52–63. doi:10.1109/tmc.2006.13. ISSN 1536-1233. S2CID 1175255.
  8. ^ Thomas M. Chen (2007). "Network Traffic Modeling". Southern Methodist University, Dallas, Texas.