Open AccessDynamic PSO for Task Scheduling Optimization in Cloud Computing
Author(s) -
M. S. Sudheer,
M Vamsi Krishna
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.b1052.0982s1119
Subject(s) - cloudsim , cloud computing , computer science , provisioning , distributed computing , particle swarm optimization , job shop scheduling , scheduling (production processes) , dynamic priority scheduling , mathematical optimization , algorithm , computer network , operating system , quality of service , routing (electronic design automation) , mathematics
Task scheduling is still a challenge in cloud computing as no existing scheduling algorithms are not effectively provisioning and scheduling the resources in the cloud. Existing authors considered only metrics like makespan, execution time and turnaround time etc. and the previous authors concentrated only to optimize the above mentioned metrics. But no existing authors were considered about the effective provisioning of the resources in the cloud i.e, compute, storage and network capacities and still many resources in the cloud were underutilized. In this paper, we want to propose an algorithm which can effectively utilize the resources in the cloud by extending Particle Swarm Optimization by addressing the metrics Bandwidth utilization and Memory utilization particularly. We have simulated this algorithm by using cloudsim and compared the modified Dynamic PSO with the PSO algorithm and it outperforms in terms of Bandwidth and Memory utilization and the makespan is also optimized.