Energy Aware Task Scheduling Algorithm In Cloud Computing Using Pso And Cuckoo Search Hybridization

Authors

  • Sudheer Mangalampalli, Vamsi Krishna Mangalampalli, Sangram Keshari Swain

Abstract

Cloud Computing is an enormous paradigm which leverages the computing power, storage and network to the users as a service on demand based on the SLA made by the Cloud user and provider. Task Scheduling is an enormous challenge in Cloud Computing as the incoming tasks on to the cloud console have to be mapped to suitable VMs which are running in the hosts at the datacenters. Mapping of the varying requests on to suitable VM is a challenging task. The objective of the task scheduling in cloud computing is to map the tasks optimally on to the virtual machines by minimizing the makespan, increasing the resource utilization there by the quality of service can be improved. In the existing literature, authors focused on makespan, resource utilization but not on energy consumption. Energy Consumption can be considered as an important parameter because it is based on consumption of the resources by the tasks so that a task scheduler is needed to optimally map the tasks on to suitable VMs by minimizing the makespan and energy consumption. In this paper, an energy aware task scheduling algorithm is developed in which the incoming tasks on to the cloud console can be mapped optimally on to the virtual machines by calculating the priority of task and priority of VM based on electricity cost at the datacenter which minimizes the makespan and energy consumption. In this paper, we have hybridized PSO and CS to solve the scheduling problem. It is implemented in the cloudsim and it is compared with the existing algorithms ACO, GA, PSO and CS. The proposed approach shows great significance over the existing algorithms in the view of makespan and energy consumption.

Published

2020-12-01

Issue

Section

Articles