Multi-Objective Optimization Techniques for Task Scheduling Problem in Distributed Systems

Abstract Task Scheduling is one of the challenging issues in distributed systems due to the allocation of multiple tasks in many processors, in order to achieve many objectives. It is known to be an NP-hard problem. These problems can be efficiently solved by population-based models. Discrete partic...

Full description

Saved in:
Bibliographic Details
Published in:Computer journal 2018-02, Vol.61 (2), p.248-263
Main Authors: Sarathambekai, S, Umamaheswari, K
Format: Article
Language:English
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:Abstract Task Scheduling is one of the challenging issues in distributed systems due to the allocation of multiple tasks in many processors, in order to achieve many objectives. It is known to be an NP-hard problem. These problems can be efficiently solved by population-based models. Discrete particle swarm optimization (DPSO) has been a recently developed population-based optimization technique which works in the discrete domain efficiently. This paper presents the DPSO variants for task scheduling problems in distributed systems to minimize the makespan, mean flow time and reliability cost. These objectives are optimized by the DPSO algorithm using the two well-known multi-objective optimization (MOO) approaches such as Aggregating and Pareto dominance. Computational simulations are done based on a set of benchmark instances to assess the performance of the MOO approaches.
ISSN:0010-4620
1460-2067
DOI:10.1093/comjnl/bxx059