A multilayer neural network for IP lookup and packet classification

IP routers use lookup tables to forward packets. They also classify packets to determine which flow they belong to in order to decide the type of quality of service they should receive. Increasing the rate of communication links and expansion of the global network is in contrast with the practical processing power of the switching devices. We propose a neural network scheme for the IP lookup problem. Our algorithm-a 12 layers neural network-represents acceptable results on the error rate and training time. Fortunately, parallel processing of neural networks provides a huge processing power to process packets. Our algorithm can be implemented in hardware on a single chip and can perform an IP lookup in only 4.5 nanoseconds implying it can support 60 Gbps link rate. Pipelining and parallel processing can be used to increase the link rate up to 400 Gbps and decrease the learning time.

[1]  Frederick J. Hill,et al.  Introduction to Switching Theory and Logical Design , 1968 .

[2]  C. Lee Giles,et al.  What Size Neural Network Gives Optimal Generalization? Convergence Properties of Backpropagation , 1998 .

[3]  Svante Carlsson,et al.  Small forwarding tables for fast routing lookups , 1997, SIGCOMM '97.

[4]  Bernhard Plattner,et al.  Scalable high speed IP routing lookups , 1997, SIGCOMM '97.

[5]  Gunnar Karlsson,et al.  IP-address lookup using LC-tries , 1999, IEEE J. Sel. Areas Commun..

[6]  Biswajit Nandy,et al.  An empirical study of today's Internet traffic for differentiated services IP QoS , 2000, Proceedings ISCC 2000. Fifth IEEE Symposium on Computers and Communications.

[7]  M. A. Jabri,et al.  Adaptive Analog VLSI Neural Systems , 1995, Springer Netherlands.

[8]  Günter Karjoth,et al.  Routing on longest-matching prefixes , 1996, TNET.

[9]  Nick McKeown,et al.  Routing lookups in hardware at memory access speeds , 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.

[10]  Lawrence T. Clark,et al.  ART1 network implementation issues , 1989, Fourth IEEE Region 10 International Conference TENCON.

[11]  R. Ravikanth,et al.  Service guarantees in the Internet: differentiated services approach , 2000, 2000 Eighth International Workshop on Quality of Service. IWQoS 2000 (Cat. No.00EX400).

[12]  N. Yazdani,et al.  Fast and scalable schemes for the IP address lookup problem , 2000, ATM 2000. Proceedings of the IEEE Conference on High Performance Switching and Routing (Cat. No.00TH8485).

[13]  Chia-Tai Chan,et al.  A fast IP routing lookup scheme , 2001, IEEE Communications Letters.

[14]  Paul Francis,et al.  Fast routing table lookup using CAMs , 1993, IEEE INFOCOM '93 The Conference on Computer Communications, Proceedings.

[15]  Anja Feldmann,et al.  Tradeoffs for packet classification , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[16]  Kohei Shiomoto,et al.  Scalable multi-QoS IP+ATM switch router architecture , 2000 .

[17]  Guru M. Parulkar,et al.  Detecting and resolving packet filter conflicts , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[18]  Kymie M. C. Tan,et al.  Detection and classification of TCP/IP network services , 1997, Proceedings 13th Annual Computer Security Applications Conference.

[19]  O. Schelen,et al.  Efficient multi-field packet classification for QoS purposes , 1999, 1999 Seventh International Workshop on Quality of Service. IWQoS'99. (Cat. No.98EX354).

[20]  George Varghese,et al.  Fast and scalable layer four switching , 1998, SIGCOMM '98.

[21]  Andrea Baraldi,et al.  A neural network for unsupervised categorization of multivalued input patterns: an application to satellite imaee clustering , 1995 .

[22]  Nick McKeown,et al.  Packet classification on multiple fields , 1999, SIGCOMM '99.

[23]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[24]  Butler W. Lampson,et al.  IP lookups using multiway and multicolumn search , 1999, TNET.

[25]  K. Singh.  A configurable 5-D packet classification engine with 4Mpacket/s throughput for high-speed data networking , 2000, 2000 IEEE International Solid-State Circuits Conference. Digest of Technical Papers (Cat. No.00CH37056).

[26]  R. G. Bennetts,et al.  Introduction to Switching Theory and Logical Design , 1981 .

[27]  Mukesh Singhal,et al.  A novel cache architecture to support layer-four packet classification at memory access speeds , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[28]  Pankaj Gupta,et al.  Algorithms for routing lookups and packet classification , 2000 .

[29]  Lex A. Akers,et al.  Integrated circuit emulation of ART1 networks , 1989 .

[30]  Bernhard Plattner,et al.  Router plugins: a software architecture for next-generation routers , 2000, TNET.