Software-defined networking (SDN) control message classification, verification, and optimization system

摘要

A traditional network reliability architecture consists of many distributed protocols, built-in dedicated network devices, and treats both data and control failures as an interrelated problem according to the physical network topologies. However, in Software-Defined Networking (SDN) architecture where the control plane is separated from the data plane, the issue of reliability is significantly related to the scalability of SDN controllers, as they are responsible for the intelligent decision of virtualization and switch policies. In this paper, we propose a novel control message classification, verification, and optimization system that serves as a fundamental approach to improve the scalability and then reliability for SDN. Unlike the traditional solutions, the proposed solution analyzes, verifies, and reduces the detrimental impact of an overloaded system by incorporating the vitality of individual control messages.