Abstract
Recent research has shown the presence of self-similarity in TCP traffic which is unaffected by the application level and human factors. This suggests the presence of protocol level contributions to network traffic self-similarity, at least in certain time scales where the effect of protocol behavior is most prominent. In this paper we show how TCP’s retransmission and congestion control mechanism contributes to the selfsimilarity of aggregate TCP flows. We develop a mathematical formulation which shows that TCP’s retransmission and congestion control mechanism results in packet dynamics of a TCP flow being analogous to a number of ON/OFF sources with OFF periods taken from a heavy tailed distribution. Using well known limit theorems, we then show that this contributes to the self-similar nature of TCP traffic. Our model shows a direct correlation of the loss rates to the degree of self-similarity. Measurements on traces collected by us also exhibit this relationship predicted by our model.
Supported in part by DARPA contract F30602-00-2-0537 and in part by DoD MURI contract F49620-97-1-0382.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Crovella, M., Bestavros, A.: Self-similarity in World Wide Web traffic: Evidence and possible causes. IEEE/ACM Trans. on Networking. 5 (1997) 835–846
Feldmann, A., Gilbert, A. C., Willinger, W.: Data networks as cascades: Investigating the multifractal nature of Internet WAN traffic. Computer Communications Review 28 (1998) 42–58
Feldmann, A., Whitt, W.: Fitting mixtures of exponentials to long-tail distributions to analyze network performance models. Proceedings of IEEE INFOCOM. (1997) 1096–1104
Feller, W. E.: An introduction to probability theory and its application. Wiley, New York (1971)
Figueiredo, D. R., Liu, B., Misra, V., Towsley, D.: On the autocorrelation structure of TCP traffic. Technical Report TR 00-55, Computer Science Department, University of Massachusetts. (2000)
Grossglauser, M., Bolot, J-C.: On the relevance of long-range dependence in network traffic. IEEE/ACM Trans. on Networking. 7 (1999) 629–640
Guo, L., Crovella, M., Matta, I.: TCP congestion control and heavy tails. Technical Report BU-CS-2000-017, Computer Science Department, Boston University. (2000)
Hill, B. M.: A simple general approach to inference about the tail of a distribution. Annals of Statistics. 3 (1975) 1163–1174
Jacobson, V.: Congestion avoidance and control. Proceedings of ACM SIGCOMM. (1988) 314–329
Kumar, A.: Comparative Performance Analysis of Versions of TCP in a Local Network with a Lossy Link. IEEE/ACM Trans. on Networking. 6 (1998) 485–498
May, M., Bonald, T., Bolot, J-C.: Analytic evaluation of RED performance. Proceedings of IEEE INFOCOM. (2000) 1415–1424
Padhye, J., Firoiu, V., Towsley D., Kurose, J.: Modeling TCP Reno performance: A simple model and its empirical validation. IEEE/ACM Trans. on Networking. 8 (2000) 133–145
Park, K., Kim, G., Crovella, M.: On the relationship between file sizes, transport protocols, and self-similar network traffic. Proceedings of International Conference on Network Protocols, (1996) 171–180
Paxson V., Floyd, S.: Wide area traffic: The failure of Poisson modeling. IEEE/ACM Trans. on Networking. 3 (1995) 226–244
Sikdar, B., Kalyanaraman, S., Vastola, K. S.: TCP Reno with random losses: Latency, throughput and sensitivity analysis. Proceedings of IEEE IPCCC, (2001) 188–195
Stevens, W. R.: TCP/IP illustrated. vol. 1. Addison Wesley (1994)
Taqqu, M. S., Willinger, W., Sherman, R.: Proof of a fundamental result in selfsimilar traffic modeling. Computer Communication Review. 27 (1997) 5–23
Taqqu, M. S., Teverovsky, V.: On estimating long-range dependence in finite and infinite variance series. In: Adler, R. J., Feldman, R. E., Taqqu, M. S., (eds.): A Practical Guide to Heavy Tails: Statistical Techniques and Applications, Birkhauser, Boston, (1998) 177–217
Veres, A., Boda, M.: The chaotic nature of TCP congestion control. Proceedings of IEEE INFOCOM. (2000) 1715–1723
Veres, A., Kenesi, Z., Molnar, S., Vattay, G.: On the Propagation of Long-Range Dependence in the Internet. Proceedings of ACM SIGCOMM. (2000) 243–254
Willinger, W., Taqqu, M. S., Sherman R., Wilson, D. V.: Self-similarity through high-variability: Statistical analysis of Ethernet LAN traffic at the source level. IEEE/ACM Trans. on Networking. 5 (1997) 71–86
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sikdar, B., Vastola, K.S. (2001). On the Contribution of TCP to the Self-Similarity of Network Traffic. In: Palazzo, S. (eds) Evolutionary Trends of the Internet. IWDC 2001. Lecture Notes in Computer Science, vol 2170. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45400-4_38
Download citation
DOI: https://doi.org/10.1007/3-540-45400-4_38
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42592-2
Online ISBN: 978-3-540-45400-7
eBook Packages: Springer Book Archive