DocumentCode :
3189992
Title :
A load balancing approach based on a genetic machine learning algorithm
Author :
Dantas, M.A.R. ; Pinto, A.R.
Author_Institution :
Dept. of Informatics & Stat., Univ. of Santa Catarina, Florianopolis, Brazil
fYear :
2005
fDate :
15-18 May 2005
Firstpage :
124
Lastpage :
130
Abstract :
Cluster configurations are a cost effective scenarios which are becoming common options to enhance several classes of applications in many organizations. In this article, we present a research work to enhance the load balancing, on dedicated and non-dedicated cluster configurations, based on a genetic machine learning algorithm. Our approach is characterized by an on time assignment scheme using a classifier system. Classifier systems are learning machine algorithms, based on high adaptable genetic algorithms. We developed a software package which was designed to test the proposed scheme in a master-slave Cow (cluster of workstation) and Now (network of workstation) environment. Experimental results, from two different operating systems, indicate the enhanced capability of our load balancing approach to adapt in cluster configurations.
Keywords :
genetic algorithms; learning (artificial intelligence); operating systems (computers); parallel processing; pattern classification; resource allocation; workstation clusters; Now; classifier system; genetic machine learning algorithm; load balancing; master-slave Cow; nondedicated cluster configurations; operating systems; workstation cluster; workstation network; Clustering algorithms; Costs; Genetic algorithms; Load management; Machine learning; Machine learning algorithms; Master-slave; Software packages; Software testing; Workstations;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High Performance Computing Systems and Applications, 2005. HPCS 2005. 19th International Symposium on
ISSN :
1550-5243
Print_ISBN :
0-7695-2343-9
Type :
conf
DOI :
10.1109/HPCS.2005.8
Filename :
1430063
Link To Document :
بازگشت