Identifying Long-Term High-Bandwidth Flows at a Router

In this paper, we propose a novel approach for identifying long-term high-rate flows at a router. Identifying the long-term high-rate flows allows a router to regulate flows intelligently in times of congestion. We employ a limited amount of state to record the arrivals of packets of individual flows. The state is managed as an LRU cache. The LRU cache self converges to hold only the flows that are high rate. The size of the cache need not depend on the number of flows at the router. We analyze a number of internet packet traces to show the effectiveness of this LRUbased cache. It is shown that this scheme is highly scalable since a few flows contribute a significant fraction of the traffic at a router.

[1]  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).

[2]  T. V. Lakshman,et al.  Design considerations for supporting TCP with per-flow queueing , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[3]  In-Koo Kim.  Analyzing network traces to identify long-term high rate flows , 2001 .

[4]  Azer Bestavros,et al.  Self-similarity in World Wide Web traffic: evidence and possible causes , 1997, TNET.

[5]  George C. Polyzos,et al.  A Parameterizable Methodology for Internet Traffic Flow Profiling , 1995, IEEE J. Sel. Areas Commun..

[6]  Sally Floyd,et al.  Promoting the use of end-to-end congestion control in the Internet , 1999, TNET.

[7]  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).

[8]  Van Jacobson,et al.  Random early detection gateways for congestion avoidance , 1993, TNET.