Tian Bu
ABSTRACT
In this paper, we explore the use of fixed point methods to evaluate the performance of a large population of TCP flows traversing a network of routers implementing active queue management (AQM) such as RED (random early detection). Both AQM routers that drop and that mark packets are considered along with infinite and finite duration TCP flows. In the case of finite duration flows, we restrict ourselves to networks containing one congested router. In all cases, we formulate a fixed point problem with the router average queue lengths as unknowns. Once these are obtained, other metrics such as router loss probability, TCP flow throughput, TCP flow end-to-end loss rates, average round trip time, and average session duration are easily obtained. Comparison with simulation for a variety of scenarios shows that the model is accurate in its predictions (mean errors less than 5%). Last, we establish monotonicity properties exhibited by the solution for a single congested router that explains several interesting observations, such as TCP SACK suffers higher loss than TCP Reno.
- 1.T. Bu, D.Towsley. "Fixed Point Approximations for TCP behavior in an AQM Network", UMass CMPSCI Technical Report 00-43Google Scholar
- 2.N. Cardwell, S. Savage, and T. Anderson. " Modeling TCP Latency " Proc. of the 2000 IEEE Infocom, Mar. 2000Google Scholar
- 3.M.E. Crovella, A. Bestavros. "Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes," IEEE/ACM Transactions on Networking, 5(6):835-846, December 1997. Google ScholarDigital Library
- 4.C. Casetti, M. Meo. "A New Approach to Model the Stationary Behavior of TCP Connections " Proc. of the 2000 IEEE Infocom, Mar. 2000Google Scholar
- 5.G. Fayolle, I. Mitrani and R. Iasnogorodski. " Sharing a Processor Among Many Job Classes" Journal of the ACM July 1980. Google ScholarDigital Library
- 6.S. Floyd, "Connection with Multiple Congested gateways in packet-Switched Networks Part 1: One-way Traffic" Computer Communication Review V.21 N.5 October 1991 Google ScholarDigital Library
- 7.K. Fall, S. Floyd. "Simulation-based Comparison of Tahoe, Reno, and SACK TCP" Computer Communication Review V. 26 N. 3 July 1996 Google ScholarDigital Library
- 8.A. Feldmann, W. Whitt. "Fitting Mixtures of exponentials to long-tail distributions to analyze network performance models" Performance evaluations 31, 1998 Google ScholarDigital Library
- 9.V. Firoiu, M. Borden. "A study of Active Queue Management for Congestion Control" Proc. of the 2000 IEEE Infocom, Mar. 2000Google Scholar
- 10.V. Firoiu, I. Yeom and X. Zhang. "A Framework for Practical Performance Evaluation and Traffic Engineering in IP Networks" Nortel Networks Technique Report, 2000Google Scholar
- 11.S. Floyd. "Notes on RED configuration," http://www.aciri.org/ oyd/red.htmlGoogle Scholar
- 12.S. Floyd. "Recommendation on using the "gentle " variant of RED," http://www.aciri.org/ oyd/red/gentle.html Mar. 2000.Google Scholar
- 13.S. Floyd, V. Jacobson. "Random early detection gateways for congestion avoidance," IEEE/ACM Trans. on Networking, 1(4), Aug. 1997. Google ScholarDigital Library
- 14.D.P. Heyman, T.V. Lakshman and A. L. Neidhardt. "A new method for analysing feedback-based protocols with applications to engineering Web traffic over the Internet," Proc. of the 1997 ACM SIGMETRICS. Google ScholarDigital Library
- 15.C.V. Hollot, V. Misra, D. Towlsey, W. Gong. "On designing improved controllers for AQM routers supporting TCP flows" To appear in Proc. of Infocom 2001Google Scholar
- 16.http://mash.cs.berkeley.edu/ns/ns.htmlGoogle Scholar
- 17.J.Mahdavi and S. Floyd. "TCP-Friendly Unicast Rate-Based Flow control," http://www.psc.edu/networking/papers/tcp friendly.htmlGoogle Scholar
- 18.A. Misra, T. Ott, J. Baras. "The window distribution of multiple TCPs with random loss queues," Proc. of Globecom'99, Dec. 1999.Google Scholar
- 19.V. Misra, W. Gong, D. Towsley. "Stochastic differential equation modeling and analysis of TCP window size behavior," Technical Report ECE-TR-CCS-99-10-01, Dept. of Electrical and Computer Engineering, Univ. of Massachusetts, Oct. 1999.Google Scholar
- 20.V. Misra, W. Gong, D. Towsley. "A Fluid-based Analysis of a Network of AQM Routers Supporting TCP Flows with an Application to RED" Proc. of ACM SIGCOMM'00, Stockholm, Sweden, September 2000. Google ScholarDigital Library
- 21.T. Ott, J. Kemperman, M. Mathis. "The stationary behavior of the ideal TCP congestion avoidance," ftp://ftp.telcordia.com/pub/tjo/TCPwindow.ps.Google Scholar
- 22.J. Padhye, V. Firoiu, D. Towsley, J. Kurose. "Modeling TCP Throughput: A Simple Model and its Empirical Validation" IEEE/ACM Transactions on Networking, 8(2), April 2000. Google ScholarDigital Library
- 23.J. Padhye, V. Firoiu and D. Towsley. "A Stochastic Model of TCP Reno Congestion Avoidance and Control." UMASS CMPSCI Technical Report 99-02, Feb 1999. Google ScholarDigital Library
- 24.L. Qiu, Y. Zhang, and S. Keshav. "On Individual and Aggregate TCP Performance" Proceedings of 7th International Conference on Network Protocols (ICNP'99), Toronto, Canada. Google ScholarDigital Library
- 25.K.K. Ramakrishnan, S. Floyd. "A proposal to add Explicit Congestion Notification (ECN) to IP," RFC 2481, January 1999. Google ScholarDigital Library
- 26.J.W. Roberts and L. Massouli. "Bandwidth Sharing and Admission Control for Elastic Traffic" ITC Specialist Seminar Yokohama, October 1998.Google Scholar
- 27.K.W. Ross. Multiservice loss networks for broadband telecommunication networks, Springer Verlag, 1995. Google ScholarDigital Library
- Fixed point approximations for TCP behavior in an AQM network
Recommendations
Fixed point approximations for TCP behavior in an AQM network
In this paper, we explore the use of fixed point methods to evaluate the performance of a large population of TCP flows traversing a network of routers implementing active queue management (AQM) such as RED (random early detection). Both AQM routers ...
Analysis of a Network of AQM Router Supporting TCP Flows and Comparison with XCP
EAIT '14: Proceedings of the 2014 Fourth International Conference of Emerging Applications of Information TechnologyThe most important issues in the Internet are congestion control. Queue management is an important part to provide better link utilization and fair service of buffer at the Internet router so as to control the congestion. To provide this we can ...
TCP CERL: congestion control enhancement over wireless networks
In this paper, we propose and verify a modified version of TCP Reno that we call TCP Congestion Control Enhancement for Random Loss (CERL). We compare the performance of TCP CERL, using simulations conducted in ns-2, to the following other TCP variants: ...
Comments