Cost Aware Load Balanced Task Scheduling with Active VM Load Evaluation

Abstract

The cloud platforms are gaining more popularity every year and adding up more customers to their portals. The cloud platforms are being flooded with the millions of user queries every second, which are becoming a major challenge to process them in the shortest possible time. The existing solutions do not evaluate the individual load on the virtual machines, while scheduling the tasks on the cloud platforms. In this paper, the new task scheduling model has been proposed, which utilizes the ant colony optimization for the load based VM allocation for each task loaded in the list for processing. The proposed model has been designed to calculate the load on the list of available VMs. The available list of the VM's is evaluated against the process cost, which checks the ability of VM in focus to process the given task. The VMs, who are eligible to process the given task, are shortlisted and the given task is assigned to the VM with the least load. The experimental results have manifested the effectiveness of the proposed model in comparison with the existing models to take the accurate offloading decisions.