Injong Rhee
Lisong Xu
Abstract
We study limitations of an equation-based congestion control protocol, called TFRC (TCP Friendly Rate Control). It examines how the three main factors that determine TFRC throughput, namely, the TCP friendly equation, loss event rate estimation and delay estimation, can influence the long-term throughput imbalance between TFRC and TCP. Especially, we show that different sending rates of competing flows cause these flows to experience different loss event rates. There are several fundamental reasons why TFRC and TCP flows have different average sending rates, from the first place. Earlier work shows that the convexity of the TCP friendly equation used in TFRC causes the sending rate difference. We report two additional reasons in this paper: (1) the convexity of 1/x where x is a loss event period and (2) different RTO (retransmission timeout period) estimations of TCP and TFRC. These factors can be the reasons for TCP and TFRC to experience initially different sending rates. But we find that the loss event rate difference due to the differing sending rates greatly amplifies the initial throughput difference; in some extreme cases, TFRC uses around 20 times more, or sometimes 10 times less, bandwidth than TCP.
- D. Bansal, H. Balakrishnan, S. Floyd, and S. Shenker. Dynamic behavior of slowly-responsive congestion control algorithms. In Proceedings of ACM SIGCOMM 2001, San Diego, CA, August 2001. Google Scholar
Digital Library
- P. Barford and M. Crovella. Generating representative web workloads for network and server performance evaluation. In Measurement and Modeling of Computer Systems, pages 151--160, 1998. Google ScholarDigital Library
- T. Bonald, M. May, and J.C. Bolot. Analytic evaluation of RED performance. In Proceedings of INFOCOM, pages 1415--1424, 2000.Google ScholarCross Ref
- K. Chen and K. Nahrstedt. Limitations of equation-based congestion control in mobile ad hoc networks. In Proc. of International Workshop on Wireless Ad Hoc Networking (WWAN 2004) in conjunction with ICDCS-2004, March 2004. Google ScholarDigital Library
- S. Floyd. HighSpeed TCP for large congestion windows. RFC 3649, December 2003. Google ScholarDigital Library
- S. Floyd, M. Handley, J. Padhye, and J. Widmer. Equation-based congestion control for unicast applications. In Proceedings of ACM SIGCOMM 2000, pages 43--56, Stockholm, Sweden, August 2000. Google ScholarDigital Library
- S. Floyd, M. Handley, J. Padhye, and J. Widmer. TCP friendly rate control (TFRC): Protocol specification. RFC 3448, January 2003. Google ScholarDigital Library
- S. Floyd and V. Jacobson. Traffic phase effects in packet-switched gateways. Internetworking:Research and Experience, 3(3):115--156, September, 1992.Google Scholar
- G. Holland and N. H. Vaidya. Analysis of TCP performance over mobile ad hoc networks. In Proceedings of IEEE/ACM MOBICOM '99, pages 219--230, August 1999. Google ScholarDigital Library
- E. Kohler, Mark Handley, Sally Floyd, and J. Padhye. Datagram congestion control protocol (DCCP). Internet Draft, draft-ietf-dccp-spec-05.txt.Google Scholar
- J. Padhye, V. Firoiu, D. Towsley, and J. Krusoe. Modeling TCP throughput: A simple model and its empirical validation. In Proceedings of the ACM SIGCOMM '98, pages 303--314, 1998. Google ScholarDigital Library
- V. Paxson. End-to-end Internet packet dynamics. IEEE/ACM Transactions on Networking, 7(3):277--292, June 1999. Google ScholarDigital Library
- V. Paxson and M. Allman. Computing TCP's retransmission timer. RFC 2988, November 2000. Google ScholarDigital Library
- PlanetLab. http://www.planet-lab.org/.Google Scholar
- I. Rhee, V. Ozdemir, and Y. Yung. TEAR: TCP emulation at receivers -- flow control for multimedia streaming. Technical report, Department of Computer Science, North Carolina State University, 2000.Google Scholar
- I. Rhee and L. Xu. Limitations of equation-based congestion control. Technical report, Department of Computer Science, North Carolina State University, 2005.Google Scholar
- M. Vojnović and J. Boudec. On the long run behavior of equation-based rate control. In Proceedings of ACM SIGCOMM 2002, pages 103--116, 2002. Also an extended version to appear in the IEEE/ACM Transactions on Networking. Google ScholarDigital Library
- J. Widmer and M. Handley. Extending equation-based congestion control to multicast applications. In Proceedings of ACM SIGCOMM 2001, San Diego, CA, August 2001. Google ScholarDigital Library
- R. Yang, M. Kim, and S. Lam. Transient behaviors of TCP-friendly congestion control protocols. In Proceedings of INFOCOM, March 2001.Google ScholarCross Ref
- Y. Zhang, N. Duffield, V. Paxson, and S. Shenker. On the constancy of Internet path properties. In Proceedings of ACM SIGCOMM Internet Measurement Workshop, November 2001. Google ScholarDigital Library
Index Terms
- Limitations of equation-based congestion control
Recommendations
Limitations of equation-based congestion control
SIGCOMM '05: Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communicationsWe study limitations of an equation-based congestion control protocol, called TFRC (TCP Friendly Rate Control). It examines how the three main factors that determine TFRC throughput, namely, the TCP friendly equation, loss event rate estimation and ...
Limitations of equation-based congestion control
We study limitations of an equation-based congestion control protocol, called TCP-Friendly Rate Control (TFRC). It examines how the three main factors that determine TFRC throughput, namely, the TCP-friendly equation, loss event rate estimation, and ...
Extending equation-based congestion control to multicast applications
Proceedings of the 2001 SIGCOMM conferenceIn this paper we introduce TFMCC, an equation-based multicast congestion control mechanism that extends the TCP-friendly TFRC protocol from the unicast to the multicast domain. The key challenges in the design of TFMCC lie in scalable round-trip time ...
Reviews
George Th. Kormentzas
Rhee and Xu examine how the three main factors that determine the transmission control protocol (TCP) friendly rate control (TFRC) throughput-the TFRC throughput equation, loss event rate estimation, and recovery time objective (RTO) estimation-can influence the long-term throughput imbalance between TFRC and TCP. They give theoretical reasons as to why such an imbalance occurs. The main findings are that any two competing flows sharing the same bottleneck link will see different loss event rates if they have significantly different sending rates, and that the loss event rate difference can greatly amplify the initial sending rate difference. To verify the theoretical findings through experiments, the paper conducts a simulation. It assumes fixed round-trip times (RTTs) and fixed packet sizes for all flows, which are not realistic assumptions. The development of an analysis that can remove these assumptions is of future interest. The paper has a clear structure, with well-organized material, a good use of language, a length suitable for its purpose, and the necessary number of appropriate references. Readers of the paper should have a basic background in protocols and the performance of systems. The paper is strongly recommended to readers and researchers dealing with new emerging networks, such as mobile ad hoc networks and high-speed long distance networks, whose characteristics are substantially different from the traditional Internet, where TCP may not work as well. For such networks, a new congestion control technique is needed, and this work can be useful for developing such a protocol. Online Computing Reviews Service
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments