A method and device for matching message features

Abstract

The invention provides a message characteristic matching device, which is applied to network equipment. The device comprises a characteristic extraction unit, a characteristic compression unit and a ternary content-addressable memory (TCAM) unit, wherein the characteristic extraction unit is used for extracting preset message characteristics from a message received from the network equipment; the characteristic compression unit is used for compressing the message characteristics; and the TCAM unit is used for searching for matching results of the message in preset access control table entries according to the compressed message characteristics, and outputting the matching results. The device can effectively save TCAM resources, fully utilizes the capacity of a TCAM and maximally develop the performance of the TCAM and simultaneously reduces the searching performance loss of each Internet protocol version 6 (IPv6) message TCAM.

Description

technical field [0001] The invention relates to network technology, in particular to a message feature matching method and device. Background technique [0002] The development of the Internet makes the application of network equipment more and more extensive. Today's network devices not only need to forward packets at Layer 2 and Layer 3, but also need to perform application processing on packets such as QoS and security. The introduction of these new features places obvious demands on various ACL matching functions for packets. In most network devices used today, the ACL technology is already very mature. Most of the general ACL lookup and matching work is done by TCAM (Ternary Content-Addressable Memory) to reduce the impact on performance. The bit width of the current mainstream TCAM can generally be configured as any one of four types: 72 bits, 144 bits, 288 bits and 576 bits. Currently, 144 bits are generally sufficient for ACL matching of IPv4 packet quintuples (so...

AI Technical Summary

Problems solved by technology

The quintuple of IPV6 is composed of 128-bit source IP, 128-bit destination IP, 16-bit source port, 16-bit destination port, and 8-bit protocol number, totaling 296 bits, so TCAM must choose 576-bit width, which is equivalent to matching The performance of each IPv6 packet is 4 times lower than that of IPv4, and the capacity is reduced to a quarter of the previous one

[0004] In addition, because each table entry will waste 576-296=280 bits of TCAM bit width, more rule matching table items are wasted, and the cost is higher; at the same time, because the bit width occupies more, the search performance of each message will be reduced. serious decline

Images