Dynamic adapting of Scalable TCP congestion control parameters

Scalable TCP is a simple sender-side alteration to the TCP congestion window update algorithm. It offers a robust mechanism to improve performance in high speed wide area networks using traditional TCP receivers. Scalable TCP uses fixed increase and decrease parameters for updating its congestion window. The performance of Scalable TCP suffers from achieving full utilization of the bandwidth when it has a long-delay connection (e.g. greater than 200msec). Also at small bandwidth delay products Scalable TCP cannot achieve full utilization whereas NewReno TCP performance is better. To overcome these limitation we propose a method to dynamically adapt the increase and decrease parameters of Scalable TCP based on the measured round trip time. Our primarily goals in design were to improve link utilization for high bandwidth and delay product networks; to improve fairness among flows with different round trip times; and to improve friendliness to NewReno for small bandwidth-delay product networks. To achieve this, we proposed a heuristic formula for the increase and decrease parameters as a function of the round trip time; we numerically evaluated and found that our proposed modifications meet our expectations.

[1]  Ren Wang,et al.  TCP westwood: Bandwidth estimation for enhanced transport over wireless links , 2001, MobiCom '01.

[2]  Róbert Szabó,et al.  Evaluation of scalable TCP , 2005, The 3rd ACS/IEEE International Conference onComputer Systems and Applications, 2005..

[3]  Manish Jain,et al.  End-to-end available bandwidth: measurement methodology, dynamics, and relation with TCP throughput , 2003, TNET.

[4]  Eitan Altman,et al.  c ○ 2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands. Fairness in MIMD Congestion Control Algorithms , 2022 .

[5]  Ren Wang,et al.  Using adaptive rate estimation to provide enhanced and robust transport over heterogeneous networks , 2002, 10th IEEE International Conference on Network Protocols, 2002. Proceedings..

[6]  Ray Jain,et al.  The art of computer systems performance analysis - techniques for experimental design, measurement, simulation, and modeling , 1991, Wiley professional computing.

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

[8]  Steven McCanne,et al.  On improving the fairness of TCP congestion avoidance , 1998, IEEE GLOBECOM 1998 (Cat. NO. 98CH36250).

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

[10]  Sally Floyd,et al.  TCP Selective Acknowledgement Options , 1996 .

[11]  Sally Floyd,et al.  TCP Selective Acknowledgment Options , 1996, RFC.

[12]  Jean C. Walrand,et al.  Analysis and comparison of TCP Reno and Vegas , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[13]  T. V. Lakshman,et al.  The performance of TCP/IP for networks with high bandwidth-delay products and random loss , 1997, TNET.

[14]  Mark Handley,et al.  Congestion control for high bandwidth-delay product networks , 2002, SIGCOMM '02.

[15]  Janey C. Hoe.  Improving the start-up behavior of a congestion control scheme for TCP , 1996, SIGCOMM 1996.

[16]  Craig Partridge,et al.  Improving round-trip time estimates in reliable transport protocols , 1991, TOCS.

[17]  V. Jacobson,et al.  Congestion avoidance and control , 1988, CCRV.

[18]  M.M. Ali,et al.  The performance of TCP congestion control algorithm over high-speed transmission links , 2004, Canadian Conference on Electrical and Computer Engineering 2004 (IEEE Cat. No.04CH37513).

[19]  Tom Kelly,et al.  Scalable TCP: improving performance in highspeed wide area networks , 2003, CCRV.