216 J. ICT Res. Appl., Vol. 15, No. 3, 2021, 216-238
Received July 8
th
, 2021, Revised September 8
th
, 2021, Accepted for publication October 4
th
, 2021.
Copyright © 2021 Published by IRCS-ITB, ISSN: 2337-5787, DOI: 10.5614/itbj.ict.res.appl.2021.15.3.2
Efficient Task Scheduling and Fair Load Distribution
Among Federated Clouds
Rajeshwari B S
1*
, M. Dakshayini
2
& H.S. Guruprasad
3
1
Department of CSE, B.M.S College of Engineering, Bangalore 560019, India
2
Department of ISE, B.M.S College of Engineering, Bangalore 560019, India
*E-mail id: madurahr@gmail.com
Abstract. The federated cloud is the future generation of cloud computing,
allowing sharing of computing and storage resources, and servicing of user tasks
among cloud providers through a centralized control mechanism. However, a great
challenge lies in the efficient management of such federated clouds and fair
distribution of the load among heterogeneous cloud providers. In our proposed
approach, called QPFS_MASG, at the federated cloud level, the incoming tasks
queue are partitioned in order to achieve a fair distribution of the load among all
cloud providers of the federated cloud. Then, at the cloud level, task scheduling
using the Modified Activity Selection by Greedy (MASG) technique assigns the
tasks to different virtual machines (VMs), considering the task deadline as the key
factor in achieving good quality of service (QoS). The proposed approach takes
care of servicing tasks within their deadline, reducing service level agreement
(SLA) violations, improving the response time of user tasks as well as achieving
fair distribution of the load among all participating cloud providers. The
QPFS_MASG was implemented using CloudSim and the evaluation result
revealed a guaranteed degree of fairness in service distribution among the cloud
providers with reduced response time and SLA violations compared to existing
approaches. Also, the evaluation results showed that the proposed approach
serviced the user tasks with minimum number of VMs.
Keywords: cloud computing; fair load distribution; federated cloud; service level
agreement; task scheduling.
1 Introduction
Federated cloud computing is a recent trend in cloud computing, where a large
cloud is formed by different cloud service providers who collaborate to provide
better cloud services [1,2]. This federated cloud is coordinated by a Federated
Cloud Broker (FCB), which interacts with the different cloud service providers
[3]. The cloud service providers make an agreement with the FCB for sharing
their resources with specific details of economical and technical aspects [4,5].
The economic as well as operational benefits of cloud computing encourage users
to send complex applications and data to the cloud [6]. There is a tremendous
growth in service requests so that the availability of resources at the right moment
and optimal distribution of requests among the federated cloud providers as well