Priority Aware Longest Job First (PA-LJF) algorithm for utilization of the resource in cloud environment
Mohit Kumar, S. C. Sharma · International Conference on Computing for Sustainable Global Development · 2016
The objective of this paper is to enhance the utilization of the resources and fulfill the user requirement, helps the load balancing in Cloud environment and provides different level of quality of service i.e. request which is coming from the user's side can be VIP or simple depends upon service selected by user. Most of the scheduling algorithm reported in the literature do not consider user's priority as like traditional Min-Min algorithm, load balancing improve Min-Min (LBIMM) algorithm, improved backfill algorithm(IBA) using balanced spiral (BS). So we need an efficient task scheduling algorithm. In this paper we have develop a task scheduling algorithm for load balancing which works dynamically and considered user priority based on Priority Aware-Longest Job First (PA-LJF) algorithm. The new develop algorithm is compare with the existing load balancing algorithm as like traditional Min-Min algorithm, LBIMM, user priority load balancing Min-Min (PA-LBIMM). It is observed that PA-LJF decrease the VIP and simple task completion time up to 20% and increase average resource utilization ratio up to 28% compare to existing algorithm. Performance of PA-LJF algorithm is slightly degrades when VIP task is increases more than 70%.