A network packet classification algorithm and system

Abstract

The invention relates to a network packet classification algorithm, comprising: dividing the existing classification rule set into a large rule subset of a source IP address and a destination IP address, a large rule subset of the source IP address, a large rule subset and a small rule subset of the destination IP address; consolidating the four equivalence classes separately; eliminating the equivalent classes with less than 5 rules in the process of merging four different subsets and storing separately in the form of linked list, and storing the addresses of the first elements in the linkedlist in the corresponding lookup tables; making priority comparison on rules from four different subsets and outputting the rules that match the packet header with the highest priority. The inventionhas the beneficial effects that the memory consumption of the lookup table construction process is reduced by combining the overall structure of the RFC algorithm and the location selection idea of the algorithm of the decision tree and combining the subset partition technology, the rule elimination technology and the location selection technology, and the high-speed packet classification processing is realized while the hardware memory validity is satisfied.

Description

technical field [0001] The invention belongs to the technical field of computer networks, and in particular relates to a network packet classification algorithm and system. Background technique [0002] With the continuous development of network technology, new network applications emerge in an endless stream. Network equipment must not only realize the basic forwarding of data packets, but also provide functions such as quality of service, traffic accounting, and differentiated services. Packet classification technology is the key technology to realize the diversification of network services, which enables network devices to process network traffic differently. However, with the surge of network traffic and continuous improvement of bandwidth, packet classification has become a bottleneck affecting network performance. [0003] Network packet classification is to compare the relevant domain values ​​of arriving data packets with the rules in the predefined classification r...

AI Technical Summary

Problems solved by technology

The hardware algorithm is mainly realized by TCAM (Ternary Content Addressable Memory, tri-state content addressable register) and FPGA (Field-Programmable Gate Array, Field Programmable Gate Array), which can achieve good performance, but its price is high and power consumption The characteristics of high density and small density greatly limit the scalability of hardware algorithms; software algorithms are mainly implemented based on general-purpose memory, which makes up for the defects of hardware algorithms. Among them, the RFC algorithm is widely used because of its high classification rate, second only to hardware algorithms. Applied to the network environment requiring high-speed lookup

At the same time, because the RFC algorithm supports parallel processing, it is very suitable for combining with hardware. However, due to its high computational and storage complexity, when the size of the rule set increases, the memory consumption of the algorithm will also rise sharply, causing serious memory expansion, making the memory Facing severe performance bottlenecks

Images