ICM: the R/sup 2/ED and edge-based congestion management

This paper proposes integrated congestion management (ICM) scheme. It is a novel domain-based congestion management scheme that can exploit the low edge-to-edge round trip times and the ingress rate regulation to achieve lower congestion recovery period (CRP). The scheme consists of two essential components. The first component is a rate-based random early detection (R/sup 2/ED) active queue management scheme providing a fair buffer allocation to individual flows in addition to an early congestion detection mechanism. The second component is the rate regulation at the ingress nodes of a domain so that lower congestion recovery periods are achieved based on the early congestion detection of the R/sup 2/ED at the core nodes. The proposed scheme does not require any per-flows state maintenance at the core nodes. Only the access nodes (first-hop) maintain per-flows rates in order to keep track of individual flow rates that are inserted into the packets. Simulation results demonstrate the several key benefits provided by the ICM scheme, including (a) fair resource allocation among individual flows (b) resilience in terms of rapid recovery from congestion (c) robustness regarding the protection of the network from aggressive sources, avoiding congestion collapse and (d) low complexity in implementation resulting in a highly scalable network. Directions for future research are further outlined.

[1]  Raj Jain,et al.  Analysis of the Increase and Decrease Algorithms for Congestion Avoidance in Computer Networks , 1989, Comput. Networks.

[2]  Konstantinos Psounis,et al.  CHOKe - a stateless active queue management scheme for approximating fair bandwidth allocation , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[3]  Raj Jain,et al.  ERICA Switch Algorithm: A Complete Description , 1996 .

[4]  Raj Jain,et al.  Congestion Avoidance in Computer Networks with a Connectionless Network Layer, Part I: Concepts, Goals and Methodology , 1988, ArXiv.

[5]  T. V. Lakshman,et al.  SRED: stabilized RED , 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).

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

[7]  Sally Floyd,et al.  A Proposal to Incorporate ECN in MPLS , 1999 .

[8]  David D. Clark,et al.  Explicit allocation of best-effort packet delivery service , 1998, TNET.

[9]  Girish Chiruvolu,et al.  An efficient edge-based congestion management for a differentiated services domain , 2000, Proceedings Ninth International Conference on Computer Communications and Networks (Cat.No.00EX440).

[10]  Robert Tappan Morris,et al.  Dynamics of random early detection , 1997, SIGCOMM '97.

[11]  Ariel Orda,et al.  QoS routing in networks with inaccurate information: theory and algorithms , 1999, TNET.

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