SDN FLOW TABLE CONFLICT DETECTION METHOD BASED ON DEEP LEARNING

摘要

Disclosed is an SDN flow table conflict detection method based on deep learning. In the method, two levels of deep learning models are used to perform conflict detection, wherein a first-level deep learning model detects whether a new flow table strategy conflicts with an existing flow table strategy, and a second-level deep learning model determines a specific flow table strategy that conflicts with the new flow table strategy in all existing flow table strategies. The present invention uses the features of abstract high-level data and automatic learning of deep learning, and compared with a traditional conflict search algorithm, can more rapidly perform conflict detection on flow table entries at a super large-scale during large-scale application deployment.