Oppositional Multi Objective Particle Swarm Based Re source Optimized Job Scheduler f or Load Balanced Cloud Service Provisioning

Abstract

Job scheduling is a key problem to be resolved in cloud service provisioning for balancing load and improving resource optimization performance. Recently, many research works have been designed for performing job scheduling using different techniques. However, job scheduling efficiency (SE) was not sufficient. In order to addresses the above limitations, Oppositional Multi-Objective Particle Swarm Based Resource Optimized Job Scheduling (OMPS-ROJS) technique is proposed. The designed OMPS-ROJS technique balances the load on computer resources by distributing tasks to available resources with higher efficiency. The OMPS-ROJS technique at first takes number of incoming user requested jobs to cloud server (CS) as input. Then, OMPS-ROJS technique develops Oppositional Particle Swarm Multi-Objective Optimization (OPSMO) algorithm in order to determine the optimal virtual machines for each input user requested jobs with a minimal amount of time. On the contrary to conventional works, OPSMO algorithm assume multi-objective such as energy, makespan, bandwidth, memory and cost for fitness function evaluation. This helps for OMPS-ROJS technique to find out the virtual machine which contains maximum resource availability as best to carry out the user requested jobs. Therefore, OMPS-ROJS technique efficientlybalancedynamic loads on CS through scheduling user requested jobs with a minimal time.Thus, OMPS-ROJS technique enhances the cloud service provisioning performance as compared to conventional works. Experimental result evident that OMPS-ROJS technique enhances the SE and lessen the EC as compared to conventional works.