Fault-Resilience for Bandwidth Management in Industrial Software-Defined Networks

Rutvij H. Jhaveri,Sagar V. Ramani,Gautam Srivastava,Thippa Reddy Gadekallu,Vaneet Aggarwal
DOI: https://doi.org/10.1109/tnse.2021.3104499
IF: 6.6
2021-10-01
IEEE Transactions on Network Science and Engineering
Abstract:Industrial Cyber-Physical Systems (ICPS) expect assurances of timely delivery of data even during the occurrence of distinct faults. It is a challenge to manage the required bandwidth by providing resilience to link failures and dynamically changing bandwidth requirements. In this paper, we address the aforementioned challenge by exploring Software-Defined Networks (SDN). We present a framework coined SDN-RMbw (Software-Defined Networking Resilience Management for Bandwidth), which is a contract-based framework, where the components are bound to bandwidth contracts and a resilience manager. The bandwidth contracts state the bandwidth requirements of traffic flows. With each such contract, a monitor is associated, which is responsible to detect two events, run-time changes and link failures. Directly after receiving the event trigger reports from the monitor, new routes are calculated by a path-finding algorithm. Based on newly calculated routes, an observer detects whether the contract requirements are still satisfied, or the contract gets violated (termed as fault). To provide resilience to such faults in the network, a resilience manager integrated with control logic decides and executes a suitable response strategy. The proposed SDN-based framework aims at providing fault-resilience as well as adapting to different network-state changes. The proposed framework is evaluated using a Ryu SDN controller on a hardware testbed. Our results show that the proposed framework provides enhanced network resilience as compared to baseline mechanisms and improves the success rate up to 21% and bandwidth up to 111 Mbps under distinct network scenarios. Furthermore, extensive experimental emulations on the Mininet tool depicts the scalability of the proposed framework.
What problem does this paper attempt to address?