Performance Analysis of Software Defined Network Concepts in Networked Embedded Systems
- PDF / 698,650 Bytes
- 13 Pages / 595.224 x 790.955 pts Page_size
- 34 Downloads / 226 Views
Performance Analysis of Software Defined Network Concepts in Networked Embedded Systems Bach Tran1 · Mohamed Elamin1 · Nghi Tran1 · Shivakumar Sastry1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Software-Defined Networking (SDN) has been largely used in wired networks to separate the control plane and the data plane and enable users to programmatically change the networking capabilities. This investigation focused on understanding how SDN concepts could be applied in a networked embedded systems environment where the nodes have limited capabilities and the wireless links have limited bandwidth. The goals were to realize the benefits of SDN while maintaining the network’s topology and connectivity. A SDN controller was implemented on BeagleBone Black Board and interfaced with a networked embedded system via a SINK node. The approach was validated through simulations and experiments based on a physical testbed in multiple scenarios. In the future, the design can be extended to a fully capable Wireless-SDN system for use in a variety of applications such as Healthcare Systems, Internet of Things and Advanced Manufacturing Systems. Keywords Wireless-SDN · Networked embedded system · Control overhead · Performance analysis
1 Introduction Network connectivity is an essential infrastructure for our economies, industries, societies and systems and has a central role in managing costs and productivity [2], and our health and wellness [3]. Modern networks confront a few challenges such as complexity and operational costs as the size of the applications increase [4]. The introduction of low-power, resource constrained nodes that are costeffective for data gathering has pushed the networking issues to their limits because of the tight bandwidth A part of the work was presented at the 5th EAI International Conference on Industrial Networks and Intelligent Systems (INISCOM), 2019 [1] Bach Tran
[email protected] Mohamed Elamin [email protected] Nghi Tran [email protected] Shivakumar Sastry [email protected] 1
Department of Electrical and Computer Engineering, University of Akron, Akron OH 44325-3904, USA
available and the unreliability of the links. Such systems are asynchronous, decentralized and severely limited, and yet they must provide reliable and trustworthy service in a variety of application settings. Software-defined networking (SDN) helps to address some of the above drawbacks by decoupling the control and data planes. With SDN, network intelligence and state are logically centralized and the underlying network infrastructure is abstracted from the applications [5]. Users are able to programmatically change the capabilities of the underlying network with minimal disruptions to the applications that rely on the networking infrastructure. A new abstraction layer is also added between applications and router/switches. SDN was primarily designed for wired networks with point-to-point (P2P) connections between their nodes [4]. On the other hand, many emerging networked sys
Data Loading...
