Makespan improvement of PSO-based dynamic scheduling in cloud environment
- 1 May 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21647054,p. 613-618
Abstract
The latest generation of distributed systems is cloud computing that has been acclaimed scientifically and commercially. CloudSim is one of the simulation tools that enables the evaluation and testing cloud services and infrastructures before development on a real cloud. For optimal use of the cloud's potential power, efficient and effective scheduling algorithms are required which can select the best resources for execution tasks. The matter of mapping and scheduling the tasks is assigning tasks to run on the existing resources in the manner that helps to maximize utilization and minimize makespan. The total time that is needed for all tasks/jobs to be finished is known as makespan. And utilization is the measure of how well the overall capacity of the cloud (network resources) is used. Due to the heterogeneity and dynamic resources, task scheduling is known as NP-complete problem and metaheuristics are needed to find the best scheduling combination. The main objective of this paper is to optimize task scheduling that uses the particle swarm optimization algorithm to minimize the makespan. Different inertia weights have been used. The Linear Descending Inertia Weight (LDIW) with an average 22.7% reduction in makespan shows the best performance.Keywords
This publication has 10 references indexed in Scilit:
- Particle swarm optimization2014
- A Dynamic Dispatching Method of Resource Based on Particle Swarm Optimization for Cloud Computing Environment2013
- A load-rebalance PSO heuristic for task matching in heterogeneous computing systems2013
- Cloud Computing—Task scheduling based on genetic algorithms2012
- A novel particle swarm optimization algorithm with adaptive inertia weightApplied Soft Computing, 2011
- A Revised Discrete Particle Swarm Optimization for Cloud Workflow Scheduling2010
- CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithmsSoftware: Practice and Experience, 2010
- A Particle Swarm Optimization-Based Heuristic for Scheduling Workflow Applications in Cloud Computing Environments2010
- Comparing with Chaotic Inertia Weights in Particle Swarm Optimization2007
- Empirical study of particle swarm optimization2003