Method for improving hardware message classification rule matching efficiency through software

Abstract

The invention provides a method for improving hardware message classification rule matching efficiency through software. Through combination of software and hardware and the hash table storage rule, the software maintains the message classification rules, adjusts collision in the hash table and puts the new rules on the head of the collision chain when the rules are added or deleted, thus easily finding the most common rules and improving the rule matching efficiency.

Description

technical field [0001] The invention relates to the field of network data management, in particular to a method for improving the matching efficiency of hardware message classification rules through software. Background technique [0002] In the hardware packet classification system, the most commonly used method for storing packet rules is to store them in the form of sequential arrays. The packet classification rules newly configured in the hardware are placed at the end of the rule table. However, due to the newly added rules Generally, it was the most commonly used at that time. As the rule capacity increases, the matching efficiency of this mechanism will decrease. [0003] In the hardware packet classification system, according to business needs, frequent operations such as adding and deleting packet classification rules are required. In the original system, the rule capacity is not very large, and it is generally stored in a sequential array. For small-capacity rules...

AI Technical Summary

Problems solved by technology

However, with the increase of the rule capacity, the prior art scheme has the disadvantage of poor matching efficiency, because the packets on the high-speed network need to quickly match the rules, and in this scheme, it takes too long to find a rule Find the entire rule array table, that is, the scale of O(n) complexity, to match the rules

Images