Traffic engineering method and system

Abstract

The invention provides a traffic engineering method, which adopts an intelligent agent, and is characterized in that a baseline module and a reinforcement learning module which are used for acquiring a traffic engineering strategy according to network topology information and service flow information are deployed in the intelligent agent; the method comprises the following steps repeatedly executed in a preset period: S1, based on network topology information and service flow information, respectively obtaining a baseline method-based traffic engineering strategy obtained by a baseline module and a reinforcement learning method-based traffic engineering strategy obtained by a reinforcement learning module through an intelligent agent; s2, comparing the security of the traffic engineering strategy based on the baseline method and the security of the traffic engineering strategy based on the reinforcement learning method, and executing the strategy with high security; and S3, storing the state information corresponding to the executed traffic engineering strategy based on the baseline method into a demonstration data set, storing the state information corresponding to the traffic engineering strategy based on the reinforcement learning method into an experience playback data set, and collecting samples from the experience playback data set and the demonstration data set to train a reinforcement learning module.

Description

technical field [0001] The present invention belongs to the technical field of traffic engineering, in particular, relates to the field of traffic engineering based on reinforcement learning in an SDN environment, and more specifically, relates to a traffic engineering method and system. Background technique [0002] With the rapid development of the Internet, the scale and complexity of the network continue to increase, and the explosive growth of network traffic puts forward higher requirements for network carrying capacity and service quality. How to optimize network traffic to ensure service quality has become the key to be solved question. [0003] The representative technology used to optimize network traffic is traffic engineering technology (Traffic Engineering, referred to as TE). The so-called traffic engineering technology refers to the purposeful scheduling and optimization of network traffic, thereby reducing network congestion and optimizing the allocation of n...

AI Technical Summary

Problems solved by technology

However, there are some problems in the existing traffic engineering method based on online reinforcement learning, which affect its use effect. Among them, the main problems include: (1) The interaction time is long, and the learning process is time-consuming

Since the online reinforcement learning method needs to collect data and update the model according to this cycle, it takes a long time for interaction to learn a good strategy, and the early effect of the learning process may not be as good as the traditional method

(2) Difficult to deploy safely

Existing methods do not judge and limit the security of the decisions generated by reinforcement learning. During the online deployment process, especially in the early stage of the learning process, it is more likely to make poor traffic engineering decisions, which will affect the network quality and even affect the network bandwidth. to negative optimization effects

Images