ISSN 2348-1196 (print) International Journal of Computer Science and Information Technology Research ISSN 2348-120X (online) Vol. 10, Issue 4, pp: (24-32), Month: October - December 2022, Available at: www.researchpublish.com
DYNAMIC LOAD BALANCING OF NETWORK RESOURCES AND DATA TRANSMISSION IN SOFTWARE-DEFINED NETWORKS Ahmad Idris Hammaadama1, Babangida Albaba Abubakar2 Department of Computer Science, Umaru Musa Yar’adua University, Katsina Nigeria DOI: https://doi.org/10.5281/zenodo.7261016
Published Date: 28-October-2022
Abstract: Traditional networking architectures have a number of important flaws that must be solved in order to satisfy today’s IT demands. Software Defined Networking (SDN) is emerging as a new networking solution to address these limitations. The implementation of a dynamic load balancing algorithm to efficiently distribute fattree network flows through multiple alternative paths between a single pair of hosts is described in this paper. The network was evaluated both before and after the load balancing algorithm was applied. The testing focused on some QoS parameters such as throughput, delay, latency, jitter, and packet loss between two servers in the fat-tree network. The performance of the proposed algorithm was evaluated through an existing result of [1], to evaluate the extent to which the maximum network resources utilization was achieved. The result obtained shows that the proposed solution was more promising in terms of minimizing end-to-end delay, latency, jitter, packet loss and maximizing throughput in which we observed a reduction in latency by 33% and packet loss by 32% while throughput was maximized by 41%. Keyword: Software-Defined Networking, Fault-tolerance, Control Plane, Controller failure.
1. INTRODUCTION Fault tolerance is a critical property for the availability of computer networks that have been studied extensively since the 1950s. However, to explore and revise new networking solutions and technologies, this concept must be revisited from time to time [2]. In this context, the newly emerging paradigm of software-defined networks (SDNs) offers a ray of hope for a new networking architecture that is more flexible and adaptable. SDN (Software-Defined Networking) is a new networking paradigm that has gained popularity in recent years. SDN decouples the control plane from the data plane [3] allowing network applications to be developed using high-level abstractions that are translated to network devices through a southbound interface [4]. In a basic SDN architecture, all of the switches in the network are controlled by a single centralized controller. It may not be ideal for the network to rely on a single controller because the controller is a single point of failure and a performance bottleneck [5]. Akyildiz et al. [6] also point out that it does not have enough control capabilities to manage the network with increasing traffic and scale. To enhance the scalability of the control plane for multi-domain SDN, several researchers propose deploying a logically centralized yet physically distributed multi-controller, such as HyperFlow [7], and Kandoo [8]. Every domain in a multi-domain network has its local controller, known as the domain controller to handle the switches and process flow requests from its domain, and domain controllers communicate with one another about their domain information to ensure a consistent network view [9]. Bari et al. [10] analyze that the controller will face a greater workload
Page | 24 Research Publish Journals