Similar articles

Abstract: The Internet evolves to incorporate very-high-bandwidth optical links and more large-delay satellite links. TCP faces new challenges in this unique environment. Theory and experiments showed that TCP becomes inefficient and is prone to be unstable as the per-flow product of bandwidth and latency increases, regardless of the queuing scheme. Variable-structure congestion control Protocol (VCP) is proposed to address these problems. However, VCP has problem in terms of convergence time, i.e., it takes a long time for a new VCP flow to achieve fair bandwidth allocation if the existing VCP flows have large congestion windows. This paper proposed an Extended Variable-structure congestion control Protocol (EVCP), which adopted a convergence controller. The basic idea of convergence controller is that if a flow has larger window than its fair window, its congestion window should be decreased more aggressively than usual in Multiplicative Decrease (MD) phase. Simulations showed that EVCP has better performance in terms of convergence time while keeping the advantages of VCP.

[1] Brakmo, L., Peterson, L., 1995. TCP Vegas: end to end congestion avoidance on a global Internet. IEEE J. Selected Areas Commun., 13(8):1465-1480.

[2] Dukkipati, N., Kobayashi, M., McKeown, N., 2005. Processor Sharing Flows in the Internet. Proc. 13th International Workshop on Quality of Service. Passau, p.286-297.

[3] Floyd, S., 2003. High Speed TCP for Large Congestion Windows. RFC 3649.

[4] Hoe, J., 1996. Improving the startup behavior of a congestion control scheme for TCP. ACM SIGCOMM Computer Commun. Rev., 26(4):270-280.

[5] Jacobson, V., 1990. Berkeley TCP Evolution from 4.3-Tahoe to 4.3 Reno. Proc. 18th Internet Engineering Task Force. University of British Colombia, Vancouver, p.365-376.

[6] Jacobson, V., Braden, R., Borman, D., 1992. TCP Extensions for High Performance. RFC 1323.

[7] Jin, C., Wei, D., Low, S., 2004. FAST TCP: Motivation, Architecture, Algorithms, Performance. Proc. INFOCOM’04. Hong Kong, p.2490-2501.

[8] Katabi, D., Handley, M., Rohrs, C., 2002. Congestion control for high bandwidth-delay product networks. ACM SIGCOMM Computer Commun. Rev., 32(4):89-102.

[9] Kelly, T., 2003. Scalable TCP: improving performance in high-speed wide area networks. ACM SIGCOMM Computer Commun. Rev., 33(2):83-91.

[10] Mathis, M., Mahdavi, J., Floyd, S., Romanow, A., 1996. TCP Selective Acknowledgement Options. RFC 2018.

[11] Nabeshima, M., Yata, K., 2004. Improving the Convergence Time of High Speed TCP. Proc. 12th IEEE International Conference on Networks. Singapore, p.19-23.

[12] Ramakrishnan, K., Floyd, S., 1999. A Proposal to Add Explicit Congestion Notification (ECN) to IP. RFC 2481.

[13] Tang, X.H., Liu, Z.L., Zhu, M.L., 2004. TCP-Rab: a receiver advertisement based TCP protocol. J. Zhejiang Univ. Sci., 5(11):1352-1360.

[14] Xia, Y., Subramanian, L., Stoica, I., Kalyanaraman, S., 2005. One More Bit is Enough. Proc. ACM SIGCOMM’05. Philadelphia, p.37-48.

[15] Xu, L., Harfoush, K., Rhee, I., 2004. Binary Increase Congestion Control (BIC) for Fast Long-Distance Networks. Proc. INFOCOM’04. Hong Kong, p.2514-2524.

[16] Zhang, Y., Leonard, D., Loguinov, D., 2006. JetMax: Scalable Max-Min Congestion Control for High-Speed Heterogeneous Networks. Proc. INFOCOM’06. Barcelona, Spain, p.1-13.