An auxiliary flow forecast control method, storage medium and electronic equipment

Abstract

The invention discloses an auxiliary traffic forecasting control method, storage medium and electronic equipment, wherein the method includes: grabbing traffic data information of links and areas within a specified topology; using the traffic data information to perform deep learning model training, Obtain a flow prediction model and a path selection model; use the flow prediction model to predict the congestion situation, and provide a sequence of services with a higher congestion probability; discard the services in the sequence whose congestion probability is less than a threshold value, and use the remaining traffic in the sequence The next business is returned to the controller; the node and link where the remaining business is located are collected separately, and the traffic trend in the real-time link of the remaining business is judged; the traffic trend is analyzed using the path selection model The business in the growth state is analyzed, and the optimal path is given; the link forwarding flow table is adjusted according to the optimal path, and the flow table is delivered to the switch and stored in the forwarding information base of the switch.

Description

technical field [0001] The invention relates to the field of network control, in particular to an auxiliary traffic forecast control method, storage medium and electronic equipment. Background technique [0002] The SDN controller is used to detect the status of the entire network, and formulate a forwarding flow table according to the protocol requirements and send it to the corresponding switch. A large amount of traffic arriving at the switch at the same stage will cause path congestion and cause network delay, which is generally avoided through traffic management. Traditional traffic management is generally carried out by network managers based on experience. With the rapid development of computer networks, network data traffic is increasing, and traditional traffic management can no longer accurately predict and control traffic. Contents of the invention [0003] In view of this, the purpose of the present invention is to propose a method, storage medium and electroni...

AI Technical Summary

Problems solved by technology

Traditional traffic management is generally carried out by network managers based on experience. With the rapid development of computer networks, network data traffic is increasing, and traditional traffic management can no longer accurately predict and control traffic.

Images