Title: Modelling and analysis of relation between load balancing and scalability in distributed computing systems

Authors: Faezeh Mollasalehi; Ehsan Mousavi Khaneghah; Zohreh Esmaeili Bidhendi

Addresses: Department of Computer Engineering, Faculty Engineering, Shahed University, Tehran, Iran ' Department of Computer Engineering, Faculty Engineering, Shahed University, Tehran, Iran ' Department of Computer Engineering, Faculty Engineering, Shahed University, Tehran, Iran

Abstract: Scalability in distributed computing systems leads to increasing the resources and changing the state of computing processes, which directly influences the functionality of load balancing. This paper presents a mathematical model to examine the relation between scalability and its impacts on the functionality of load balancing based on vector algebra, the vectorial definition of global computing processes, mathematical definition of inhomogeneity, and definition of vectors describing the state of the resource and the process, as well as using the concept of scalability given load balancing. By using this model, it would be possible for system management to decide whether scalability can enhance the efficiency of distributed computing systems or not. The conducted experiments have indicated that the presented model can describe the impacts of scalability on the functionality of load balancing.

Keywords: scalability; dynamical load balancing; distributed exascale computing systems; cluster computing; grid computing; vector algebra; global activity.

DOI: 10.1504/IJCSYSE.2021.123562

International Journal of Computational Systems Engineering, 2021 Vol.6 No.6, pp.281 - 298

Received: 29 Aug 2021
Accepted: 15 Jan 2022

Published online: 27 Jun 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article