Network Optimization for Safety-Critical Systems Using Software-Defined Networks

Perner, Cora

doi:10.1007/978-3-319-77610-1_10

Cited by 2 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this scenario it is difficult to motivate the use of Smart Routing. Several studies like those in References [16][17][18] have addressed the problem of optimising failure restoration as an Integer Linear Program (ILP). However ILP-based approaches may be slow to converge in large-scale networks [19].…”

Section: Related Workmentioning

confidence: 99%

Rapid Restoration Techniques for Software-Defined Networks

2020

View full text Add to dashboard Cite

There is increasing demand in modern day business applications for communication networks to be robust and reliable due to the complexity and critical nature of such applications. As such, data delivery is expected to be reliable and secure even in the harshest of environments. Software-Defined Networking (SDN) is gaining traction as a promising approach for designing network architectures which are robust and flexible. One reason for this is that separating the data plane from the control plane, increases the controller’s ability to configure the network rapidly. When network failure events occur, the network manager may trade-off the optimality of the achieved network reconfiguration with the responsivenss of the reconfiguration process. Responsiveness may be favoured when the network resources are under stress and the failure rate is high. We contribute SDN recovery methods that leverage information about the structure of the network to expedite network restoration when a link failure occurs. They operate by detecting community-like structures in the network topology and then they find alternative paths which have low operation and installation costs using this information. Extensive simulations are conducted to evaluate the proposed SDN recovery methods using open-source simulation tools. They provide evidence that the proposed approaches lead to performance gains when an alternative path is required among a set of candidate paths.

show abstract