Method and system for preventing message congestion

Abstract

The invention relates to a method and a system for preventing message congestion. The system comprises a processor, which is connected with a traffic manager and used for completing message classification, parsing, modification and encapsulation. Congestion processing is carried out according to the received congestion strategy sent by the traffic manager. The traffic manager is used for the detection of message congestion and the generation of the congestion strategy and used for sending the congestion strategy to a packet processor. According to the method and the system of the invention, the message processing performance and efficiency are improved.

Description

technical field [0001] The invention relates to the field of data communication, in particular to a message congestion avoidance method and system. Background technique [0002] With the rapid development of the network, the rapid growth of network equipment in terms of technology and performance, the IP services carried on the Internet (network) are also gradually diversified, and comprehensive services such as rich data, voice, and video are booming. With the continuous increase of network bandwidth from 10Mbps, 100Mbps, 1Gbps, 10Gbps to 100Gbps, relying on suppliers to blindly increase bandwidth and expand network equipment investment is increasingly insufficient to meet the requirements. End-to-end QOS (Quality of Service) management tools have gradually become the focus of suppliers. According to different customers and different business requirements for network resources, different service qualities are provided, such as providing dedicated bandwidth, reducing messag...

AI Technical Summary

Problems solved by technology

[0013] The main problem with the first implementation is: the packet processor is used to classify packet flows, the traffic manager is used to avoid congestion, and the traffic manager is located downstream of the packet processor.

When discarding due to congestion, packets must be discarded in the traffic manager, which wastes some resources compared to discarding packets in the packet processor

[0014] The main problem with the second implementation method is: congestion avoidance is performed directly on the packet processor, not only the flow classification of the packet must be completed on the packet processor, but also the congestion detection of the packet needs to be completed, and the congestion of the packet Detection means that a large number of queues need to be implemented and maintained in the packet processor software, and discard policy processing

For packet processors, implementing and maintaining a large number of queues brings additional memory overhead, and requires multiple accesses to peripherals, and its performance and efficiency are not guaranteed

Images