Packet classification algorithm using multiple subspace intersecting

Authors:
Mingfeng Tan;Zexin Lu;Lei Gao
Affiliations:
School of Computer Science, National University of Defense Technology, Changsha, China;School of Computer Science, National University of Defense Technology, Changsha, China;School of Computer Science, National University of Defense Technology, Changsha, China
Venue:
ICCNMC'05 Proceedings of the Third international conference on Networking and Mobile Computing
Year:
2005

Citing 5
Cited 0

Fast and scalable layer four switching

Proceedings of the ACM SIGCOMM '98 conference on Applications, technologies, architectures, and protocols for computer communication
High-speed policy-based packet forwarding using efficient multi-dimensional range matching

Proceedings of the ACM SIGCOMM '98 conference on Applications, technologies, architectures, and protocols for computer communication
Packet classification on multiple fields

Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Classifying Packets with Hierarchical Intelligent Cuttings

IEEE Micro
Packet classification using multidimensional cutting

Proceedings of the 2003 conference on Applications, technologies, architectures, and protocols for computer communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Packet classification on multi-fields is difficult and has poor worst-case performance due to its character of multiple dimensions. Thus his paper proposes a efficient hardware algorithm MSI (multiple subspace intersecting) to solve it. MSI cuts each dimension of the classifier into several subspaces and then utilizes the parallelism of hardware to do the classification on these subspaces. For a classifier with n rules of width W, MSI needs only less than n[4W+log2(n)] bits. MSI is able to classify 100 M packets/s with pipelined hardware, supports fast incremental update, and has good flexibility in specification of rule and. By simulation, we find MSI has better performance comparing to some existing algorithms.