An improved load balanced metaheuristic scheduling in cloud

Cloud computing refers to on-demand delivery of service over internet and has application in various domains like media, research, business, bigdata analysis etc. Task scheduling is one of the prime issues in this type of environment. Various metaheuristic algorithms and hard optimization problems h...

Full description

Saved in:
Bibliographic Details
Published in:Cluster computing 2019-09, Vol.22 (Suppl 5), p.10873-10881
Main Authors: Aruna, M., Bhanu, D., Karthik, S.
Format: Article
Language:English
Subjects:
Algorithms
Cloud computing
Communication
Computer Communication Networks
Computer Science
Efficiency
Energy consumption
Heuristic methods
Infrastructure
Load balancing
Operating Systems
Particle swarm optimization
Polynomials
Power
Processor Architectures
Response time
Software
Task scheduling
Workloads
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cloud computing refers to on-demand delivery of service over internet and has application in various domains like media, research, business, bigdata analysis etc. Task scheduling is one of the prime issues in this type of environment. Various metaheuristic algorithms and hard optimization problems have been proposed for solving cloud task scheduling which is a non-deterministic polynomial or an NP. Adaptation of the scheduling strategy to the changes taking place in the environment has to be done by a good scheduler. A proposal for cloud scheduling by means of a balanced load using both firefly algorithm (FA) and particle swarm optimization (PSO) heuristics has been made. The aim is to balance the load of the entire system while at the same time bring down the makespan of a set of tasks. This new strategy for scheduling has been simulated with CloudSim tool kit package. The results of this experiment proved that the proposed FA performed better than min–min scheduling, PSO, and also the first come first serve methods.
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-017-1213-9