A HighSpeed TCP Study : Characteristics and Deployment Issues

The current congestion control mechanism used in TCP has difficulty reaching full utilization on high speed links, particularly on wide-area connections. For example, the packet drop rate needed to fill a Gigabit pipe using the present TCP protocol is below the currently achievable fiber optic error rates. HighSpeed TCP was recently proposed as a modification of TCP’s congestion control mechanism to allow it to achieve reasonable performance in high speed wide-area links. In this paper, simulation results showing the performance of HighSpeed TCP and the impact of its use on the present implementation of TCP are presented. Network conditions including different degrees of congestion, different levels of loss rate, different degrees of bursty traffic and two distinct router queue management policies were simulated. The performance and fairness of HighSpeed TCP were compared to the existing TCP and solutions for bulk-data transfer using parallel streams.

[1]  Masayuki Murata,et al.  Survey on Fairness Issues in TCP Congestion Control Mechanisms , 2001 .

[2]  Evandro de Souza,et al.  A simulation-based study of HighSpeed TCP and its deployment , 2003 .

[3]  David L. Black,et al.  The Addition of Explicit Congestion Notification (ECN) to IP , 2001, RFC.

[4]  S. Shenker,et al.  Observations on the dynamics of a congestion control algorithm: the effects of two-way traffic , 1991, SIGCOMM '91.

[5]  Sally Floyd,et al.  HighSpeed TCP for Large Congestion Windows , 2003, RFC.

[6]  Van Jacobson,et al.  Traffic phase effects in packet-switched gateways , 1991, CCRV.

[7]  Van Jacobson,et al.  TCP Extensions for High Performance , 1992, RFC.

[8]  Brian D. Noble,et al.  The Effects of Systemic Packet Loss on Aggregate TCP Flows , 2002, ACM/IEEE SC 2002 Conference (SC'02).

[9]  B. Barden.  Recommendations on queue management and congestion avoidance in the Internet , 1998 .

[10]  Jason Lee,et al.  Microscopic examination of TCP flows over transatlantic links , 2003, Future Gener. Comput. Syst..

[11]  JasonLee,et al.  Applied Techniques for High Bandwidth Data Transfers across Wide Area Networks , 2001 .

[12]  Sally Floyd,et al.  Simulation-based comparisons of Tahoe, Reno and SACK TCP , 1996, CCRV.

[13]  Steven H. Low,et al.  Stabilized Vegas , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[14]  Wu-chun Feng,et al.  DYNAMIC RIGHT-SIZING IN TCP. , 2001 .

[15]  G. Fairhurst,et al.  Performance limitations due to TCP Burstiness in GEO satellite networks with limited buffering , 2000 .

[16]  Matthew Mathis,et al.  Automatic TCP buffer tuning , 1998, SIGCOMM '98.

[17]  FloydSally,et al.  Simulation-based comparisons of Tahoe, Reno and SACK TCP , 1996 .

[18]  Fernando Paganini,et al.  FAST TCP: from theory to experiments , 2005, IEEE Network.

[19]  Mark Handley,et al.  Congestion control for high bandwidth-delay product networks , 2002, SIGCOMM.

[20]  Sally Floyd.  Limited Slow-Start for TCP with Large Congestion Windows , 2004, RFC.

[21]  Robert T. Braden,et al.  Requirements for Internet Hosts - Communication Layers , 1989, RFC.

[22]  T. V. Lakshman,et al.  Performance Analysis of Window-based Flow Control Using TCP/IP: Effect of High Bandwidth-Delay Products and Random Loss , 1994, High Performance Networking.

[23]  K. K. Ramakrishnan,et al.  A Proposal to add Explicit Congestion Notification (ECN) to IP , 1999, RFC.