System and method for classifying network messages on basis of hybrid computation hardware

Abstract

The invention discloses a system and a method for classifying network messages on basis of hybrid computation hardware. According to the system and the method, centralized dispatching is carried out on various different computation hardware resources by a central processor to construct a multistage classified processing assembly line, matching rules are reasonably divided according to the characteristics of different computation hardware and configured in each stage of the assembly line, classified processing for simple rules can be finished by a special hardware chip, and complex custom classifying rules can be cooperatively realized by the special hardware chip, a universal parallel processor and a universal central processor, thus enhancing the capacity of processing messages.

Description

technical field [0001] The invention relates to a multi-level network message classification system and method based on hybrid computing hardware, belonging to the technical field of message classification. Background technique [0002] When a computer network device receives an external network message from a network interface, it needs to process the message according to its own tasks. For example, routers and switches need to forward the message to other network interfaces after receiving it, and the NAT gateway device receives the message. After the data message, the message needs to be modified, and the load balancer needs to redirect the message. Before operating (forwarding or modifying) the packets, these devices must first classify the data packets, and perform different operations for different types of network packets. The existing technology usually uses a dedicated hardware chip to extract the content of the fixed position of the message and compares it with ea...

AI Technical Summary

Problems solved by technology

The traditional method of using dedicated hardware chips to classify packets has insufficient flexibility. Users can only select from a few given rules such as the destination IP address and physical address of network packets, and it is difficult to classify according to user-defined message fields.

New network packet classification methods try to use general-purpose processors or graphics processors to implement packet classification to achieve a flexibly defined packet classification mechanism. However, existing methods are limited to a single type of computing hardware (CPU or GPU). Its packet classification processing capability is very limited, and it is difficult to meet the classification processing requirements of massive network packets in existing computer networks

Images