DocumentCode :
2742522
Title :
A New Method for Scheduling Load Balancing in Multi-processor Systems Based on PSO
Author :
Kiyarazm, OmidReza ; Moeinzadeh, M-Hossein ; Sharifian-R, Sarah
Author_Institution :
Kashmar Branch, Dept. of Comput. Eng., Islamic Azad Univ., Kashmar, Iran
fYear :
2011
fDate :
25-27 Jan. 2011
Firstpage :
71
Lastpage :
76
Abstract :
In parallel and distributed systems, optimal task scheduling and load balancing has always been of great interest in order to minimize the time and speed up the process which plays a major role in the efficiency of such systems. By load balancing we mean scheduling the jobs in a way that every job could be executed concurrently while it is mapped to a processing unit, such as a processor (in a multi-processor system) or a computer (in a grid computer). It has been proved that dynamic load balancing is an NP-complete problem and designing an optimal scheduler that increases the efficiency and speed of the parallel systems is desired. In this paper, we want to present an algorithm for task scheduling and load balancing in multi-processor systems based on PSO method that can minimize the maximum span and average utilization of all processors in an optimal way.
Keywords :
computational complexity; particle swarm optimisation; processor scheduling; resource allocation; NP-complete problem; distributed system; dynamic load balancing; multiprocessor system; optimal scheduler; optimal task scheduling; parallel system; particle swarm optimisation; Algorithm design and analysis; Computers; Heuristic algorithms; Load management; Optimal scheduling; Processor scheduling; Scheduling; dynamic load balancingt; multi-processor systems; particle swarm optimization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Systems, Modelling and Simulation (ISMS), 2011 Second International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4244-9809-3
Type :
conf
DOI :
10.1109/ISMS.2011.73
Filename :
5730323
Link To Document :
بازگشت