2-level Hierarchical Search Method Based on tcam for Range Multi-Matching

Abstract

The invention proposes a TCAM-based range multi-matching two-level hierarchical search method, and realizes port range multi-matching in the TCAM-based packet classification problem. Its characteristic is that first extracts the wide interval range from the range set, and processes the wide interval range separately, and then, for the remaining range set, merges as many related ranges as possible to form some New range sets, and most of these recombined new ranges are composed of multiple original ranges. Then, for new ranges that contain more than one original range among these new ranges, divide them into several contiguous ranges with a mutual distance of 1. In the search process, first search the merged new range in the first-level TCAM+SRAM, and then determine whether it is necessary to search the second-level TCAM+SRAM according to the search results of the new range set, so that The TCAM-based range multi-matching technology is realized through these two divisional searches.

Description

technical field [0001] The present invention relates to a TCAM (Ternary Content Addressable Memory, three-state content addressable memory)-based two-level hierarchical search method for solving the multi-matching problem of ranges, which is mainly used in network devices such as intrusion detection systems, routers, and switches. The invention belongs to the technical field of network data packet classification in computer network. Background technique [0002] With the rapid development of network technology, especially the emergence of various new network applications in recent years, such as Network Intrusion Detection System (Network Intrusion Detection System, NIDS), Quality of Service (Quality of Service, QoS), Load Balance (Load Balance) etc. need to report all matching results or partial matching results. The traditional software-based packet classification technology is far from meeting the requirements of wire speed, but hardware-based technology can solve this pr...

AI Technical Summary

Problems solved by technology

[0004] TCAM can be used in packet classification technology mainly because it has the advantages of fast and accurate search speed, general structure and simple operation, but it also has several obvious disadvantages: high cost, high power consumption and complicated routing update

However, this method has disadvantages such as low search efficiency, the need to use extension bits, occupying additional TCAM space, low update efficiency, and the need to preprocess search keywords.

Chinese Patent: A Prefix Extension Method for Triple Content Addressable Memory Range Matching, Application No. 200610011517, Publication No. 101039253B, and Chinese Patent: Segmented Encoding Extension Method for Triple Content Addressable Memory Range Matching, Application No. 200610011465 , Publication No. 101035061B, proposed a prefix extension method and a segmented code extension method to solve the TCAM range matching problem, but these two methods still need several prefixes to replace a range, therefore, they cannot reduce the storage of TCAM very well In addition, if both the source port and the destination port are in the range, since the expansion number of the prefix is ​​the multiplication of the expansion number of the two port prefixes, a rule requires more TCAM entries to store, so it does not Applicable when the rule set contains many fields that contain multiple scopes at the same time

In addition, this paper does not use the idea of ​​TCAM partition search, but only searches all range sets in TCAM, so the search power consumption is much larger than that of the present invention

Images