Abstract:Software-defined networking (SDN) has emerged as a new network paradigm that promises control/data plane separation and centralized network control. While these features simplify network management and enable innovative networking, they give rise to persistent concerns about <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">reliability</italic> . The new paradigm suffers from the disadvantage that various network faults may consistently undermine the reliability of such a network, and such faults are often new and difficult to resolve with existing solutions. To ensure SDN reliability, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">fault management</italic> , which is concerned with detecting, localizing, correcting and preventing faults, has become a key component in SDN networks. Although many SDN fault management solutions have been proposed, we find that they often resolve SDN faults from an incomplete perspective which may result in side effects. More critically, as the SDN paradigm evolves, additional fault types are being exposed. Therefore, comprehensive reviews and constant improvements are required to remain on the leading edge of SDN fault management. In this paper, we present the first comprehensive and systematic survey of SDN faults and related management solutions identified through advancements in both the research community and industry. We apply a systematic classification of SDN faults, compare and analyze existing SDN fault management solutions in the literature, and conduct a gap analysis between solutions developed in an academic research context and practical deployments. The current challenges and emerging trends are also noted as potential future research directions. This paper aims to provide academic researchers and industrial engineers with a comprehensive survey with the hope of advancing SDN and inspiring new solutions.

Predictive Fault Management in the Dynamic Environment of Ip Networks

Research on Network Management Systems Via Multi-Agent Technology

Probabilistic Inference for Network Management

An Inference Algorithm for Probabilistic Fault Management in Distributed Systems

Application of Innovation Grey Theory to Fault Predication of Uncertainty Systems

Dynamic Bayesian Network for Robust Latent Variable Modeling and Fault Classification

THE BAYESIAN NETWORKS METHOD FOR NETWORK FAULT MANAGEMENT

A New Fault Detection Method for Computer Networks

A Fault Prognosis Strategy Based on Time-Delayed Digraph Model and Principal Component Analysis

Network Management of Predictive Mobile Networks

A dynamic prognosis algorithm in distributed fault tolerant model predictive control

Backward Inference in Bayesian Networks for Distributed Systems Management

Fault prediction approach for dynamic system

Fault Diagnosis of Train Network Control Management System Based on Dynamic Fault Tree and Bayesian Network

Network Traffic Prediction and Fault Detection Based on Adaptive Linear Model

Fault Management in Software-Defined Networking: A Survey

Fault Prediction for Dynamic Systems with State-Dependent Faults

An adaptive decision method using structure feature analysis on dynamic fault propagation model

A Prediction-Based Dynamic Resource Management Approach for Network Virtualization.

Multi-Fault Diagnosis for Industrial Processes Based on Hybrid Dynamic Bayesian Network

A Fault Tolerance Scheme based on Packet-loss Fault Prediction