Journal of Theoretical & Computational Science
Open Access

Abstract

A Multi-Objective Clustered Input Oriented Salp Swarm Algorithm in Cloud Computing

Juliet A Murali, Brindha T

The Infrastructure as a Service (IaaS) cloud computing service model facilitates on-demand sharing of computing resources over the internet. The scheduling of these resources is the main dispute that has to be handled. Recently, different swarm-based algorithms have been implemented for scheduling. The main intention of the cloud scheduling problem is to satisfy user requirements by allocating resources. This paper proposed a two-phase task scheduling algorithm named Clustered Input Oriented Salp Swarm Algorithm (CIOSSA). In the first phase, named Task Splitting Agglomerative Clustering (TSAC), the tasks are categorized based on the deadline and the output generated by TSAC becomes the input to the next phase. The Input Oriented Salp Swarm Algorithm (IOSSA) is used for the scheduling process. It is a multi-objective version of the fundamental Salp Swarm Algorithm (SSA) which experience slow convergence rate. Here a fundamental version SSA is proposed with two different objectives satisfied by two separate leaders. The use of two leaders instead of one expands the search space of the optimization problem. This proposed algorithm achieved efficient utilization of resources, which in turn reduced the cost. It also improves the schedule make span. The experimental analysis is performed in cloudsim. The simulation result shows that the proposed CIOSSA framework will produce better results than the existing algorithm

Published Date: 2024-09-11; Received Date: 2024-08-12