A network routing and forwarding method and system based on business service quality requirements

Abstract

The invention discloses a network routing and forwarding method and system based on business service quality requirements, belonging to the field of network communication. The present invention uses time delay, jitter, and bandwidth constraints to express service quality requirements. Compared with the prior art, it more accurately matches business services, expresses business requirements, refines and distinguishes different types of services, and realizes differentiated requirements for diversified services. Embed specific service quality binding indicators in BPP metadata. The forwarding node directly understands the specific service quality index requirements by extracting the BPP header information, so as to customize the routing strategy and reward standard according to the business service quality demand index, obtain the optimal routing configuration under this requirement, and realize the differentiated requirements of diversified services . The DDPG method based on deep reinforcement learning is adopted, and the learning target is the weight of each link. Using the trained link weights, OSPF and other routing algorithms are used to select routes, and the optimal network routing strategy is obtained to improve network utility.

Description

technical field [0001] The invention belongs to the technical field of network communication, and more specifically relates to a method and system for network routing and forwarding based on business service quality requirements. Background technique [0002] With the vigorous development of the Internet and the emergence of new technologies, network applications have changed from single data transmission to diversified network application scenarios, and traditional networks are also unable to support the multi-dimensional and differentiated network requirements of new Internet formats. In the Industrial Internet, business service quality is directly related to production safety and public safety. Different application scenarios have different emphasis on service response time, delay tolerance, service security level, and service reliability. For example, in precision instrument manufacturing, etc. In industrial scenarios, the bearer requirements for low latency, low jitter,...

AI Technical Summary

Problems solved by technology

[0003] Traditional routing solutions face many difficulties. The network architecture and protocol design have limitations, and there are problems such as slow convergence speed of routing algorithms and difficulty in providing network transmission guarantees for different services.

Most of the traffic-based routing algorithms used in the past use the shortest path algorithm measured by the number of hops, without considering the business requirements for delay, bandwidth, jitter, etc., not only did not make full use of network resources, select the optimal path, but Unable to learn from past experience and customize personalized network forwarding strategies for different business types

Inefficient routing decisions can also lead to excessive loading of network links, increasing end-to-end transmission delays

Images