Design of a High Speed FPGA-Based Classifier for Efficient Packet Classification

Abstract

Packet classification is a vital and complicated task as the processing of packets should be done at a specified line speed. In order to classify a packet as belonging to a particular flow or set of flows, network nodes must perform a search over a set of filters using multiple fields of the packet as the search key. Hence the matching of packets should be much faster and simpler for quick processing and classification. A hardware accelerator or a classifier has been proposed here using a modified version of the HyperCuts packet classification algorithm. A new pre-cutting process has been implemented to reduce the memory size to fit in an FPGA. This classifier can classify packets with high speed and with a power consumption factor of less than 3W. This methodology removes the need for floating point division to be performed by replacing the region compaction scheme of HyperCuts by pre-cutting, while classifying the packets and concentrates on classifying the packets at the core of the network.