Supratim Deb
ABSTRACT
Mathematical models of congestion control capture the congestion indication mechanism at the router in two different ways: rate-based models, where the queue-length at the router does not explicitly appear in the model, and queue-based models, where the queue length at the router is explicitly a part of the model. Even though most congestion indication mechanisms use the queue length to compute the packet marking or dropping probability to indicate congestion, we argue that, depending upon the choice of the parameters of the AQM scheme, one would obtain a rate-based model or a rate-and-queue-based model as the deterministic limit of a stochastic system with a large number of users. We also consider the impact of implementing AQM schemes in the real queue or a virtual queue. If an AQM scheme is implemented in a real queue, we show that, to ensure that the queuing delays are negligible compared to RTTs, one is forced to choose the parameters of a AQM scheme in a manner which yields a rate-based deterministic model. On the other hand, if the AQM scheme is implemented in a virtual queue, small-queue operation is achieved independent of the choice of the parameters, thus showing a robustness property of virtual queue-based schemes.
- T. Alpcan and T. Başar. A utility-based congestion control scheme for internet-style networks with delay. In Proceedings of IEEE Infocom, pages 2039--2048, San Francisco, California, March-April 2003.Google Scholar
Cross Ref
- E. Altman, K. Avrachenkov, C. Barakat, and R. Nunez-Queija. State-dependent M/G/1 type queueing analysis for congestion control in data networks. Computer Networks, 39(6):789--808, August 2002. Google ScholarDigital Library
- S. Athuraliya, D. E. Lapsley, and S. H. Low. Random early marking for Internet congestion control. In Proceedings of IEEE GLOBECOM, pages 1747--1752, Rio de Janeiro, Brazil, 1999.Google Scholar
- S. Athuraliya and S. H. Low. Optimization flow control, I: Basic algorithm and convergence. IEEE/ACM Transactions on Networking, pages 861--875, December 1999. Google ScholarDigital Library
- S. Athuraliya and S. H. Low. Optimization flow control, II: Implementation. 2000. http://netlab.caltech.edu/.Google Scholar
- F. Baccelli and D. Hong. Flow level simulation of large IP networks. In Proceedings of INFOCOM, pages 1911--1921, April 2003.Google ScholarCross Ref
- H. Chen and D. D. Yao. Fundamentals of Queueing Networks. Springer-Verlag, New York, 2000.Google Scholar
- D. J. Daley and D. Vere-Jones. An Introduction to the Theory of Point Processes. Springer-Verlag, New York, 1988.Google Scholar
- S. Deb. Deterministic fluid models for internet congestion control. PhD thesis, University of Illinois at Urbana-Champaign, September 2003. Google ScholarDigital Library
- S. Deb, S. Shakkottai, and R. Srikant. Stability and convergence of TCP flows in a many-flows regime. In Proceedings of INFOCOM, pages 884--894, San Francisco, CA, April.Google Scholar
- S. Deb, and R. Srikant. Stochastic models of proportionally fair congestion controllers. In Proceedings of IEEE Conference on Decision and Control, Dec 2003.Google ScholarCross Ref
- S. Floyd. TCP and explicit congestion notification. ACM Computer Communication Review, 24:10--23, October 1994. Google ScholarDigital Library
- S. Floyd and V. Jacobson. Random early detection gateways for congestion avoidance. IEEE/ACM Transactions on Networking, 1(4):397--413, August 1993. Google ScholarDigital Library
- R. J. Gibbens and F. P. Kelly. Resource pricing and the evolution of congestion control. Automatica, 35:1969--1985, 1999. Google ScholarDigital Library
- C. V. Hollot and Y. Chait. Nonlinear stability analysis for a class of TCP/AQM schemes. In Proceedings of the IEEE Conference on Decision and Control, pages 2309--2314, Orlando, FL, December 2001.Google Scholar
- C. V. Hollot, Y. Liu, V. Mishra, and D. Towsley. Unresponsive flows and AQM performance. In Proceedings of INFOCOM, volume 1, pages 85--95, San Francisco, CA, April 2003.Google ScholarCross Ref
- C. V. Hollot, V. Misra, D. Towsley, and W. Gong. On designing improved controllers for AQM routers supporting TCP flows. In Proceedings of INFOCOM, pages 1726--1734, Anchorage, Alaska, April 2001.Google ScholarCross Ref
- F. P. Kelly. Models for a self-managed Internet. Philosophical Transactions of the Royal Society, A358:2335--2348, 2000.Google ScholarCross Ref
- F. P. Kelly. Fairness and stability of end-to-end congestion control. European Journal of Control, 9:149--165, 2003.Google ScholarCross Ref
- F. P. Kelly, A. Maulloo, and D. Tan. Rate control in communication networks: shadow prices, proportional fairness and stability. Journal of the Operational Research Society, 49:237--252, 1998.Google ScholarCross Ref
- L. Kleinrock. Queueing Systems, Volume 1: Theory. Wiley--Interscience, New York, 1975. Google ScholarDigital Library
- S. Kunniyur and R. Srikant. Analysis and design of an adaptive virtual queue algorithm for active queue management. In Proceedings of ACM Sigcomm, pages 123--134, 2001. Google ScholarDigital Library
- S. Kunniyur and R. Srikant. A time-scale decomposition approach to adaptive ECN marking. In Proceedings of IEEE INFOCOM, pages 1330--1339, Anchorage, Alaska, April 2001.Google ScholarCross Ref
- S. Kunniyur and R. Srikant. End-to-end congestion control: utility functions, random losses and ECN marks. IEEE/ACM Transactions on Networking, to be published. An earlier version appeared in Proceedings of IEEE INFOCOM, 2000. Google ScholarDigital Library
- S. Liu, T. Başar, and R. Srikant. Controlling the Internet: A survey and some new results. In Proceedings of IEEE Conference on Decision and Control, Maui, Hawaii, December 2003.Google Scholar
- S. H. Low, F. Paganini, J. Wang, S. Adlakha, and J. C. Doyle. Dynamics of TCP/RED and a scalable control. In Proceedings of INFOCOM, pages 239--248, New York, June 2002.Google ScholarCross Ref
- V. Misra, W. Gong, and D. Towsley. Stochastic differential equation modeling and analysis of TCP windowsize behavior. In Proceedings of Performance'99, Istanbul, Turkey, October 1999.Google Scholar
- T. J. Ott. ECN protocols and TCP paradigm. http://web.njit.edu/~ott/Papers, 1999.Google Scholar
- S. Shakkottai and R. Srikant. Mean FDE models for Internet congestion control. 2002,. http://www.comm.csl.uiuc.edu/~srikant. A shorter version appeared in the Proceedings of IEEE INFOCOM, 2002.Google Scholar
- P. Tinnakornsrisuphap and A. Makowski. Limit behavior of ECN/RED gateways under a large number of TCP flows. In Proceedings of INFOCOM, pages 873--883, San Francisco, CA, April 2003.Google ScholarCross Ref
- G. Vinnicombe. On the stability of networks operating TCP-like congestion control. In Proceedings of the IFAC World Congress, Barcelona, Spain, 2002. Also, University of Cambridge Technical Report CUED/F-INFENG/TR.398, http://www.eng.cam.ac.uk/gv.Google ScholarCross Ref
- J. T. Wen and M. Arcak. A unifying passivity framework for network flow control. In Proceedings of IEEE Infocom, pages 1156--1166, San Francisco, California, April 2003.Google ScholarCross Ref
- Y. Yi, S. Deb, and S. Shakkottai. Short queue regime and rate based marking. In Proceedings of Conference on Information Science and Systems, Princeton, NJ, March 2004.Google Scholar
Index Terms
- Rate-based versus queue-based models of congestion control
Recommendations
Rate-based versus queue-based models of congestion control
Mathematical models of congestion control capture the congestion indication mechanism at the router in two different ways: rate-based models, where the queue-length at the router does not explicitly appear in the model, and queue-based models, where the ...
A Rate-Based End-to-End Multicast Congestion Control Protocol
ISCC '00: Proceedings of the Fifth IEEE Symposium on Computers and Communications (ISCC 2000)Current reliable multicast protocols do not have scalable congestion control mechanisms and this deficiency leads to concerns that multicast deployment may endanger stability of the network. In this paper, we present a sender-based approach for ...
Advances in Active Queue Management (AQM) Based TCP Congestion Control
Current end-to-end Internet congestion control under tail-drop (TD) queue management experiences performance degradations such as multiple packet losses, high queueing delay and low link utilization. In this paper, we review recently proposed active ...
Comments