Modelling, Design and Optimization of Dependable Softwarized Networks for Industrial Applications

Abstract

The recent trend of Industry 4.0 promotes the concepts of "industrial internet and digital factory", which requires the transformation of local industrial networks, as well as wide area networks carrying industrial traffic. Softwarized network architectures, i.e., Software Defined Networking (SDN) and Network Function Virtualization (NFV), can aid this transition by providing a fine grained network traffic control and high degree of programmability, with open standards and protocol stack. While first studies and prototypes have demonstrated the feasibility of SDN/NFV based networks, a comprehensive study of dependability of softwarized networks is still missing in state of the art literature. Recent studies on reliability in softwarzed network have focused mainly on the network failures, while the reliability of software functions has been widely overlooked or oversimplified. Our research aims to close this gap, by providing the contribution in three following areas: i) evaluation and forecasting of the reliability of control plane functions, ii) characterization of failure dynamics between control and data plane, and iii) design of robust wide area networks in the presence of software and network failures.