Network flow load balancing control method based on reinforcement learning

Abstract

The invention discloses a network flow load balancing control method based on reinforcement learning, which comprises the following steps of: 1) selecting an action ai with a maximal return value from an action set of the next hop according to the state quantity s and the strategy pi of a current data packet when the data packet is in a routing node R*; 2) modifying the state quantity s of the data packet according to actual conditions of the data packet after the current data packet is routed, and updating the action set of the next hop of the current data packet; 3) modifying the rewards and punishment values r of the current data packet according to balancing states of the current network flow; and 4) updating the strategy pi according to the rewards and punishment values; and repeating the step 1) to the step 4) when the current data packet reaches a final destination address. According to the method, optimal or approximately optimal control on load balancing of the network flow is realized by unceasing interactive learning of an intelligent agent and the network environment.

Description

Technical field [0001] The invention relates to the technical field of network traffic load balancing, and specifically refers to providing an intelligent network traffic load balancing control method based on reinforcement learning. Background technique [0002] With the rapid development of the network, various network applications are emerging one after another, and the traffic on the corresponding network is also increasing. According to authoritative surveys, network quality of service (Qos: Quality of Service), especially network response time, is the main factor that affects user experience. Therefore, reasonable network design and network service quality are issues that every network engineer needs to consider . There are many solutions to ensure the quality of service of the network. The present invention proposes a network traffic load balancing algorithm based on reinforcement learning. Data packets can use the reinforcement learning algorithm to select a suitable for...

AI Technical Summary

Problems solved by technology

The RIP algorithm is too simple, poor in security, and consumes a lot of bandwidth, so it is not suitable for large-scale networks; the OSPF algorithm is a link-state routing protocol. Compared with the RIP algorithm, it has the advantages of fast convergence, low protocol overhead, high security, and wide adaptability. Its configuration is complicated, and its routing load balancing ability is weak; the EIGRP algorithm is an enhanced gateway internal routing protocol, which has many advantages, but it belongs to Cisco's private protocol

Images