ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Load Balancing in Software-Defined Networks using Adaptive Generic Master and Slave Architecture

Journal: International Journal of Mechanical and Production Engineering Research and Development (IJMPERD ) (Vol.10, No. 3)

Publication Date:

Authors : ; ;

Page : 3439-3454

Keywords : SDN; Load Balancer; Load Balancing Algorithms; Advance Generic Controller Adaptive Load Balancing (AGCALB); Switches & Controller;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Fault Tolerance is a major and integral parameter of network strength and flexibility. The systems and mechanisms that follow fault tolerance are expected to make sure the reliability, availability and flexibility of a network at a very high level at several platforms. Introductions of Software-Defined Networking (SDN) has open new ways to develop new layouts, standards, parameters and architectures in the favour of fault tolerance. In this paper, the two architectures are represented and Fault Tolerance is carried out on these two respective architectures: (1) Centralized master controller consisting four slave controllers. (2) several slave controllers. The model proposed is called Adaptive Load Balancing Controller (AGCALB) It balances the load among slave controllers using heuristic algorithm. Tool used for simulation phase is mininet. Controller taken into account is floodlight controller. Jitter, delay, throughput and response time are used to check the performance. AGCALB is compared with two existing models : (1) Hyperflow (Kreutz et. al., 2012) and (2) ECFT (Aly and Al-anasi, 2018). The results obtained are quite promising, the AGCALB throughput is increased by 16%, jitter and delay decreased by 14%, and 15% respectively, and their is a better response of 13%, when compared to Hyperflow and when compared to ECFT throughput increased by 19%, jitter and delay decreased by 10% and 17% respectively and response time is better by 15%.

Last modified: 2021-01-06 13:59:48