An analytical RED function design guaranteeing stable system behavior

This paper introduces the procedure for an analytical derivation of the drop probability function for the random early detection queue management mechanism. The procedure is based on a model of the TCP window system stability with methods of control theory. The term stability refers to the oscillation amplitude of the steady state queue size. The outcome is a polynomial function of the queue size, which can be approximated with a power function of the queue size. Besides the goal of achieving low oscillations for the whole load range, this derived function shows yet other advantages. There are no parameters to adjust, which extends the applicability of RED to a wide range of load situations applying the derived function. Additionally, the under-utilization at low loads, as often experienced with the original RED function, is avoided automatically. Simulations show clear evidence that the derived function outperforms the original linear RED function.

[1]  Donald F. Towsley,et al.  A control theoretic analysis of RED , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[2]  Deborah Estrin,et al.  Recommendations on Queue Management and Congestion Avoidance in the Internet , 1998, RFC.

[3]  Paul Barford,et al.  Generating representative Web workloads for network and server performance evaluation , 1998, SIGMETRICS '98/PERFORMANCE '98.

[4]  Kang G. Shin,et al.  A self-configuring RED gateway , 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).

[5]  Sally Floyd,et al.  TCP and explicit congestion notification , 1994, CCRV.

[6]  Victor Firoiu,et al.  A study of active queue management for congestion control , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[7]  QUTdN QeO,et al.  Random early detection gateways for congestion avoidance , 1993, TNET.

[8]  Thomas Ziegler.  On averaging for active queue management congestion avoidance , 2002, Proceedings ISCC 2002 Seventh International Symposium on Computers and Communications.

[9]  Thomas Ziegler,et al.  On the Non-Linearity of the RED Drop Function , 2002 .

[10]  Sally Floyd,et al.  Adaptive RED: An Algorithm for Increasing the Robustness of RED's Active Queue Management , 2001 .

[11]  Vishal Misra,et al.  Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED , 2000, SIGCOMM.

[12]  Serge Fdida,et al.  Comparison of tail drop and active queue management performance for bulk-data and Web-like Internet traffic , 2001, Proceedings. Sixth IEEE Symposium on Computers and Communications.

[13]  Donald F. Towsley,et al.  On designing improved controllers for AQM routers supporting TCP flows , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).