IP网拥塞控制技术研究
|
摘要
基于IP的Internet是采用分组交换的网络，同时使用网络的用户数量和他们的负载都没有限制，因此很容易导致拥塞。为了防止拥塞，人们提出了一些拥塞控制方案。目前，IP网中采用的拥塞控制包括终端控制机制和链路控制机制两个有机部分。本文分别对它们进行了概括性的研究介绍，并将侧重点放在端到端拥塞控制的研究上。
针对可靠业务传输的TCP拥塞控制协议是目前互联网中最主要的端到端拥塞控制协议。深入研究TCP拥塞控制协议不仅可以更有效地提高网络效率，同时也可以为研究其他类型的拥塞控制提供参考和标准，保证Internet在加入其他协议后仍能健康、稳定地运行。本文研究了TCP协议拥塞控制的实现算法，包括慢启动、拥塞避免、快速恢复和快速重传。然后较为详细地介绍了两种TCP吞吐量模型：简单模型和复杂模型，并给出了它们的详细推导过程。
随着Internet的发展，出现了很多新的诸如实时多媒体流之类的技术和基于这些新技术的商业应用。这些应用由于自身的原因是采用对拥塞没有反应的UDP作为传输协议的，这将带来拥塞崩溃和不公平的危险。为了解决这个问题，人们提出了很多适用于实时业务，同时保持对TCP公平性的拥塞控制协议，本文对这些拥塞控制机制进行了介绍。重点研究了一种基于吞吐率复杂模型的TCP友好拥塞控制机制——TFRC。详细分析了该机制的基本原理、报文格式、丢包率测量和速率调整方式等细节问题，并通过仿真实验，验证其在保持对TCP的友好性的同时，体现出比TCP更适用实时业务传输的性能。
无线通信和网络技术的发展对于传输协议中的拥塞控制研究提出了新的挑战．由于无线通信自身特点，如远远高于有线介质的衰减，干扰，切换等现象，使得无线数据通信中数据包丢失的原因已不仅仅局限于拥塞，因此基于丢包事件的传统拥塞控制机制就不能正确反映网络的实际．本文研究了无线网络环境中的TCP拥塞控制策略和实时业务拥塞控制策略，提出了一种新的无线TCP友好速率控制机制。
Internet is a packet-switched network .It is prone to congestion since neither the number of users nor their workloads are limited. In the past years ,a number of congestion control schemes have been proposed .Today, the congestion control used in Internet consists of two implementations :end point algorithm and link algorithm .Both of them are roughly studied in the thesis .We focus on the end-to-end congestion control in our research.
TCP congestion control is the primary congestion control mechanisms used in nowadays Internet. Deeply studying TCP congestion control can not only improve the efficiency of network but also provide the reference and standard to other congestion control mechanisms. In this thesis, we study the basic algorithm of TCP congestion control including slow start 、 congestion avoidance 、 fast recovery and fast retransmit ,and then we introduce two TCP throughput models :simple model and complex model in detail.
With the development of Internet, more and more real-time applications are applied. To fulfill their demands for communication, these applications are transported by UDP, which can bring the problem of unfairness and congestion collapse. To solve this problem, many TCP friendly congestion control mechanisms are proposed. We discuss a TCP throughput model based TCP-friendly congestion control
mechanism------TFRC, analyzing its basic theory、 packet format、 measurement of loss
rate and sending rate adjustment. The simulation shows its better performance to real-time application over TCP as well as keeping the friendliness to TCP.
The development of wireless communication and network brings new challenge on the congestion control research. Because of its unique characteristics such as fading, interferes, handoffs, etc, the cause of packet loss can not be narrowed down on the congestion. The traditional congestion control mechanism based on packet loss detection is not suitable for the reality of the current network environment. In this paper, we focus our research on the TCP congestion control and real-time application congestion control in wireless environment. New algorithms are proposed and analyzed.
针对可靠业务传输的TCP拥塞控制协议是目前互联网中最主要的端到端拥塞控制协议。深入研究TCP拥塞控制协议不仅可以更有效地提高网络效率，同时也可以为研究其他类型的拥塞控制提供参考和标准，保证Internet在加入其他协议后仍能健康、稳定地运行。本文研究了TCP协议拥塞控制的实现算法，包括慢启动、拥塞避免、快速恢复和快速重传。然后较为详细地介绍了两种TCP吞吐量模型：简单模型和复杂模型，并给出了它们的详细推导过程。
随着Internet的发展，出现了很多新的诸如实时多媒体流之类的技术和基于这些新技术的商业应用。这些应用由于自身的原因是采用对拥塞没有反应的UDP作为传输协议的，这将带来拥塞崩溃和不公平的危险。为了解决这个问题，人们提出了很多适用于实时业务，同时保持对TCP公平性的拥塞控制协议，本文对这些拥塞控制机制进行了介绍。重点研究了一种基于吞吐率复杂模型的TCP友好拥塞控制机制——TFRC。详细分析了该机制的基本原理、报文格式、丢包率测量和速率调整方式等细节问题，并通过仿真实验，验证其在保持对TCP的友好性的同时，体现出比TCP更适用实时业务传输的性能。
无线通信和网络技术的发展对于传输协议中的拥塞控制研究提出了新的挑战．由于无线通信自身特点，如远远高于有线介质的衰减，干扰，切换等现象，使得无线数据通信中数据包丢失的原因已不仅仅局限于拥塞，因此基于丢包事件的传统拥塞控制机制就不能正确反映网络的实际．本文研究了无线网络环境中的TCP拥塞控制策略和实时业务拥塞控制策略，提出了一种新的无线TCP友好速率控制机制。
Internet is a packet-switched network .It is prone to congestion since neither the number of users nor their workloads are limited. In the past years ,a number of congestion control schemes have been proposed .Today, the congestion control used in Internet consists of two implementations :end point algorithm and link algorithm .Both of them are roughly studied in the thesis .We focus on the end-to-end congestion control in our research.
TCP congestion control is the primary congestion control mechanisms used in nowadays Internet. Deeply studying TCP congestion control can not only improve the efficiency of network but also provide the reference and standard to other congestion control mechanisms. In this thesis, we study the basic algorithm of TCP congestion control including slow start 、 congestion avoidance 、 fast recovery and fast retransmit ,and then we introduce two TCP throughput models :simple model and complex model in detail.
With the development of Internet, more and more real-time applications are applied. To fulfill their demands for communication, these applications are transported by UDP, which can bring the problem of unfairness and congestion collapse. To solve this problem, many TCP friendly congestion control mechanisms are proposed. We discuss a TCP throughput model based TCP-friendly congestion control
mechanism------TFRC, analyzing its basic theory、 packet format、 measurement of loss
rate and sending rate adjustment. The simulation shows its better performance to real-time application over TCP as well as keeping the friendliness to TCP.
The development of wireless communication and network brings new challenge on the congestion control research. Because of its unique characteristics such as fading, interferes, handoffs, etc, the cause of packet loss can not be narrowed down on the congestion. The traditional congestion control mechanism based on packet loss detection is not suitable for the reality of the current network environment. In this paper, we focus our research on the TCP congestion control and real-time application congestion control in wireless environment. New algorithms are proposed and analyzed.
引文
[1] Postel, J. Transmission Control Protocol. RFC 793, 1981.
[2] Jacobson, V. Congestion avoidance and control. ACM Computer Communication Review, 1988, 18(4): 314-329.
[3] William Stallings. High-Speed Networks and Internets Performance and Quality of Service. 2nd edition. Prentice Hall 2002.
[4] Tanenbaum, A. S. Computer Networks. 3rd edition., Prentice Hall, Inc., 1996.
[5] Jain, R., Ramakrishnan, K. K., Chiu, Dah-Ming. Congestion avoidance in computer networks with a connectionless network layer. Technical Report, DEC-TR-506, Digital Equipment Corporation, 1988. http://www.cis.ohio-state./jain.
[6] Peterson, L. L., Davie, B. S. Computer Networks: a System Approach. Morgan Kaufmann Publishers, 2000.
[7] Bennett, J. C. R., Partridge, C., Shectman, N. Packet reordering is not pathological network behavior. IEEE/ACM Transactions on Networking, 1999, 7(6): 789-798.
[8] Shenker, S. Fundamental design issues for the future Internet. IEEE Journal on Selected Areas in Communications, 1995, 13(7): 1176-1188.
[9] Jain, R. Congestion control in computer networks: issues and trends. IEEE Network Magazine, 1990, 4(3): 24-30.
[10] 罗万明 林闯 阎保平．TCP／IP拥塞控制研究．计算机学报 2001 24(1)：1-18．
[11] Allman M, Paxson V, Stevens W. TCP Congestion Control. Request for Comments: 2581, April 1999
[12] Hoe, J. C. Improving the start-up behavior of a congestion control scheme for TCP. Computer Communication Review, 1996, 26(4): 270-280.
[13] Zhang, Y., Qiu, L., Keshav, S. Optimizing TCP start-up performance. Technical Report, CS Department, Cornell University, 1999. http://www.cs.cornell.edu/cnrg/papers.html.
[14] Allman, M., Floyd, S., Partridge, C. Increasing TCP's Initial Window. RFC 2414, 1998.
[15] Wang, H., Xin, H., Reeves, D. S., et al. A simple refinement of slow-start of TCP congestion control. In: Sherif, S. H., ed. Proceedings of the IEEE Symposium on Computers and Communications (ISCC 2000). Los Alamitos, CA: IEEE Computer Society, 2000. 98-105.
[16] Padmanabhan, V.N. Addressing the challenges of Web data transport [Ph.D. Thesis]. Berkeley, CA: University of California,Berkeley, 1998.
[17] Visweswaraiah, V., Heidemann, J. Improving restart of idle TCP connections. Technical Report, USC TR 97-661, University of Southern California, 1997. http:// www.isi.edu/johnh/PAPERS/.
[18] Ke, J. Willianmson, C. Towards a rate-based TCP protocol for the Web. In: Boukerche, A., ed. Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS 2000).San Fransisco, CA: IEEE Computer Society, 2000. 36-45.
[19] Aggarwal, A., Savage, S., Anderson, T., Understanding the performance of TCP pacing. In: Sidi, M., ed. Proceedings of the IEEE INFOCOM. Tel Avlv: IEEE Communications Society, 2000. 1157-1165.
[20] Floyd, S. Connections with Multiple Congested Gateways in Packet-Switched Networks Part 2: Two-way Traffic. 1991. http://www. aciri.org/floyd.
[21] Border, J., Kojo, M., Griner, J., et al. Performance Enhancing Proxies Intended to Mitigate Link-Related Degradations. RFC 3135, 2001.
[22] Jacobson, V. Congestion avoidance and control. ACM Computer Communication Review, 1988,18(4):314-329.
[23] Ludwig, R., Katz, R.H. The Eifel algorithm: making TCP robust against spurious retransmissions. ACM Computer Communication Review, 2000,30(1):23-36.
[24] Ramakrishnan, K.K., Jain, R. A binary feedback scheme for congestion avoidance in computer networks. ACM Transactions on Computer Systems, 1990,8(2): 158-181.
[25] Floyd, S. TCP and explicit congestion notification. ACM Computer Communication Review, 1994, 24(5): 10-23.
[26] Floyd, S., Handley, M., Padhye, J., et al. Equation-Based congestion control for Unicast Applications. ACM Computer Communication Review, 2000,30(4):43-56.
[27] Widmer, J., Denda, R., Mauve, M. A survey on TCP-friendly congestion control. IEEE Network, 2001,15(3):28-37.
[28] Padhye, J., Firoiu, V., Towsley, D.F., et al. Modeling TCP Reno performance: a simple model and its empirical validation. IEEE/ACM Transactions on Networking, 2000,8(2): 133-45.
[29] Balakrishnan, H., Padmanabhan, V.N., Seshan, S., et al. A comparison of mechanisms for improving TCP performance over wireless links. IEEE/ACM Transactions on Networking, 1997,5(6):756-769.
[30] Partridge, C, Shepard, T. J. TCP/IP performance over satellite links. IEEE Network, 1997,11(5):44-49.
[31] Kalampoukas, L., Varma, A., Ramakrishnan, K.K. Improving TCP throughput over two-way asymmetric link: analysis and solutions. Performance Evaluation Review, 1998,26(1):78-89.
[32] S.Floyd, K.Fall. Promoting the Use of End-to-End Congestion Control in the Internet. IEEE/ACM Transactions on Networks,Aug.l999.7(4):458-472
[33] J.Padhye, V.Firoiu, D.Towsley. Modeling TCP Throughput: A Simple Model and its Empirical Validation. SIGCOMM Symposium on Communications Architectures and Protocols, Aug. 1998:303-304.
[34] S. Bodamer, "A New Scheduling Mechanism to Provide Relative Differentiation for Real-Time IP Traffic," Proc.GLOBECOM, 2000, vol. 1,pp. 646-50.
[35] Floyd, S., Handley, M., Padhye, J. and J. Widmer, "Equation-Based Congestion Control for Unicast Applications", August 2000,Proc.ACM SIGCOM 2000.
[36] Floyd, S ., Handley, M., Padhye, J. and J. Widmer,"TCP Friendly Rate Control (TFRC):Protocol Specification",January 2003,Network Working Group RFC 3448.
[37] Padhye, J., Firoiu, V., Towsley, D. and J. Kurose, "Modeling TCP Throughput: A Simple Model and its Empirical Validation",Proc. ACM SIGCOMM 1998.
[38] Volkan Ozdemir and Injong Rhee," TCP Emulation At Receivers (TEAR)," presentation slides. Presented at IRTF, RM meeting, Technical report, 2000
[39] Floyd, S ., Handley, M., Padhye, J. and J. Widmer, "Equation-Based Congestion Control for Unicast Applications: the Extended Version",ICSI tech report TR-00-03,March 2000.
[40] Paxson, V. End-to-End routing behavior in the Internet. IEEE/ACM Transactions on Networking, 1997,5(5):601-615.
[41] Seshan, S., Stemm, M., Katz, R.H. Network measurement architecture for adaptive applications. In: Sidi, M., ed. Proceedings of the IEEE INFOCOM. Tel Avlv: IEEE Communications Society, 2000. 285-294.
[42] Balakrishman, H., Seshan, S., Stemm, M., et al. Analyzing stability in wide-area network performance. Performance Evaluation Review, 1997, 25(1): 2-12.
[43] Paxson, V. Measurements and analysis of end-to-end Intemet dynamics [Ph. D. Thesis]. Berkeley, CA: University of California, Berkeley, 1996.
[2] Jacobson, V. Congestion avoidance and control. ACM Computer Communication Review, 1988, 18(4): 314-329.
[3] William Stallings. High-Speed Networks and Internets Performance and Quality of Service. 2nd edition. Prentice Hall 2002.
[4] Tanenbaum, A. S. Computer Networks. 3rd edition., Prentice Hall, Inc., 1996.
[5] Jain, R., Ramakrishnan, K. K., Chiu, Dah-Ming. Congestion avoidance in computer networks with a connectionless network layer. Technical Report, DEC-TR-506, Digital Equipment Corporation, 1988. http://www.cis.ohio-state./jain.
[6] Peterson, L. L., Davie, B. S. Computer Networks: a System Approach. Morgan Kaufmann Publishers, 2000.
[7] Bennett, J. C. R., Partridge, C., Shectman, N. Packet reordering is not pathological network behavior. IEEE/ACM Transactions on Networking, 1999, 7(6): 789-798.
[8] Shenker, S. Fundamental design issues for the future Internet. IEEE Journal on Selected Areas in Communications, 1995, 13(7): 1176-1188.
[9] Jain, R. Congestion control in computer networks: issues and trends. IEEE Network Magazine, 1990, 4(3): 24-30.
[10] 罗万明 林闯 阎保平．TCP／IP拥塞控制研究．计算机学报 2001 24(1)：1-18．
[11] Allman M, Paxson V, Stevens W. TCP Congestion Control. Request for Comments: 2581, April 1999
[12] Hoe, J. C. Improving the start-up behavior of a congestion control scheme for TCP. Computer Communication Review, 1996, 26(4): 270-280.
[13] Zhang, Y., Qiu, L., Keshav, S. Optimizing TCP start-up performance. Technical Report, CS Department, Cornell University, 1999. http://www.cs.cornell.edu/cnrg/papers.html.
[14] Allman, M., Floyd, S., Partridge, C. Increasing TCP's Initial Window. RFC 2414, 1998.
[15] Wang, H., Xin, H., Reeves, D. S., et al. A simple refinement of slow-start of TCP congestion control. In: Sherif, S. H., ed. Proceedings of the IEEE Symposium on Computers and Communications (ISCC 2000). Los Alamitos, CA: IEEE Computer Society, 2000. 98-105.
[16] Padmanabhan, V.N. Addressing the challenges of Web data transport [Ph.D. Thesis]. Berkeley, CA: University of California,Berkeley, 1998.
[17] Visweswaraiah, V., Heidemann, J. Improving restart of idle TCP connections. Technical Report, USC TR 97-661, University of Southern California, 1997. http:// www.isi.edu/johnh/PAPERS/.
[18] Ke, J. Willianmson, C. Towards a rate-based TCP protocol for the Web. In: Boukerche, A., ed. Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS 2000).San Fransisco, CA: IEEE Computer Society, 2000. 36-45.
[19] Aggarwal, A., Savage, S., Anderson, T., Understanding the performance of TCP pacing. In: Sidi, M., ed. Proceedings of the IEEE INFOCOM. Tel Avlv: IEEE Communications Society, 2000. 1157-1165.
[20] Floyd, S. Connections with Multiple Congested Gateways in Packet-Switched Networks Part 2: Two-way Traffic. 1991. http://www. aciri.org/floyd.
[21] Border, J., Kojo, M., Griner, J., et al. Performance Enhancing Proxies Intended to Mitigate Link-Related Degradations. RFC 3135, 2001.
[22] Jacobson, V. Congestion avoidance and control. ACM Computer Communication Review, 1988,18(4):314-329.
[23] Ludwig, R., Katz, R.H. The Eifel algorithm: making TCP robust against spurious retransmissions. ACM Computer Communication Review, 2000,30(1):23-36.
[24] Ramakrishnan, K.K., Jain, R. A binary feedback scheme for congestion avoidance in computer networks. ACM Transactions on Computer Systems, 1990,8(2): 158-181.
[25] Floyd, S. TCP and explicit congestion notification. ACM Computer Communication Review, 1994, 24(5): 10-23.
[26] Floyd, S., Handley, M., Padhye, J., et al. Equation-Based congestion control for Unicast Applications. ACM Computer Communication Review, 2000,30(4):43-56.
[27] Widmer, J., Denda, R., Mauve, M. A survey on TCP-friendly congestion control. IEEE Network, 2001,15(3):28-37.
[28] Padhye, J., Firoiu, V., Towsley, D.F., et al. Modeling TCP Reno performance: a simple model and its empirical validation. IEEE/ACM Transactions on Networking, 2000,8(2): 133-45.
[29] Balakrishnan, H., Padmanabhan, V.N., Seshan, S., et al. A comparison of mechanisms for improving TCP performance over wireless links. IEEE/ACM Transactions on Networking, 1997,5(6):756-769.
[30] Partridge, C, Shepard, T. J. TCP/IP performance over satellite links. IEEE Network, 1997,11(5):44-49.
[31] Kalampoukas, L., Varma, A., Ramakrishnan, K.K. Improving TCP throughput over two-way asymmetric link: analysis and solutions. Performance Evaluation Review, 1998,26(1):78-89.
[32] S.Floyd, K.Fall. Promoting the Use of End-to-End Congestion Control in the Internet. IEEE/ACM Transactions on Networks,Aug.l999.7(4):458-472
[33] J.Padhye, V.Firoiu, D.Towsley. Modeling TCP Throughput: A Simple Model and its Empirical Validation. SIGCOMM Symposium on Communications Architectures and Protocols, Aug. 1998:303-304.
[34] S. Bodamer, "A New Scheduling Mechanism to Provide Relative Differentiation for Real-Time IP Traffic," Proc.GLOBECOM, 2000, vol. 1,pp. 646-50.
[35] Floyd, S., Handley, M., Padhye, J. and J. Widmer, "Equation-Based Congestion Control for Unicast Applications", August 2000,Proc.ACM SIGCOM 2000.
[36] Floyd, S ., Handley, M., Padhye, J. and J. Widmer,"TCP Friendly Rate Control (TFRC):Protocol Specification",January 2003,Network Working Group RFC 3448.
[37] Padhye, J., Firoiu, V., Towsley, D. and J. Kurose, "Modeling TCP Throughput: A Simple Model and its Empirical Validation",Proc. ACM SIGCOMM 1998.
[38] Volkan Ozdemir and Injong Rhee," TCP Emulation At Receivers (TEAR)," presentation slides. Presented at IRTF, RM meeting, Technical report, 2000
[39] Floyd, S ., Handley, M., Padhye, J. and J. Widmer, "Equation-Based Congestion Control for Unicast Applications: the Extended Version",ICSI tech report TR-00-03,March 2000.
[40] Paxson, V. End-to-End routing behavior in the Internet. IEEE/ACM Transactions on Networking, 1997,5(5):601-615.
[41] Seshan, S., Stemm, M., Katz, R.H. Network measurement architecture for adaptive applications. In: Sidi, M., ed. Proceedings of the IEEE INFOCOM. Tel Avlv: IEEE Communications Society, 2000. 285-294.
[42] Balakrishman, H., Seshan, S., Stemm, M., et al. Analyzing stability in wide-area network performance. Performance Evaluation Review, 1997, 25(1): 2-12.
[43] Paxson, V. Measurements and analysis of end-to-end Intemet dynamics [Ph. D. Thesis]. Berkeley, CA: University of California, Berkeley, 1996.