Link state prediction method and system for sdn network based on service flow information

Abstract

The invention discloses a link state prediction method based on service flow information in an SDN network, which includes: obtaining data flow in the link; counting the data flow according to the protocol to obtain a statistical index; obtaining a calculation index according to the statistical index; To predict the link state, the present invention obtains the statistical index by counting the data flow according to the agreement, and then obtains the calculation index including the service information through the statistical index, and uses the calculation index as the input of the prediction model for the final prediction, because the calculation index includes Service category information (such as the amount of data transmitted by each service and the proportion of different services, etc.), makes the input granularity update, thus effectively improving the accuracy of link state prediction.

Description

Technical field [0001] The present invention belongs to the field of SDN network technology, in particular to an SDN network based on service flow information link state prediction method and system. Background [0002] Software-defined networks (Sofeware Defined Networks) have a globally controllable control plane and an efficient forwarding plane, and decision makers can schedule and optimize network resources based on global information. Global information includes the network topology and the link state of each link in the topology (such as the used link bandwidth, latency, packet loss rate, etc.) of the link, and SDN needs to realize accurate scheduling in a timely and accurate sense of link state. [0003] The current link state perception methods mainly measure and predict. The former uses active measurement or passive measurement to obtain the link state under the current network, while the latter analyzes the change trend of traffic on the link and predicts the link sta...

AI Technical Summary

Problems solved by technology

However, because the current service flow information measurement method can only obtain service flow information at the end nodes through active packet sending or plug-in monitoring, it is difficult for the intermediate nodes of the network to passively obtain upper-layer application information. The existing neural network-based link state prediction methods only Rely on information below the application layer for prediction, the input is still at a coarse granularity, and the prediction accuracy is not high

Images