|2/2004 - 10|
Using Agent Technology Combined with ACO Metaheuristics on Load Balancing AlgorithmsMitica CRAUS, Catalin BULANCEA
|Click to see author's profile on SCOPUS, IEEE Xplore, Web of Science|
|Not available online | Views: 662|
load balancing, agent, Ant Colony Optimisation, metaheuristic
References keywords will be displayed on the next page reload.
About this article
Date of Publication: 2004-04-02
Volume 4, Issue 2, Year 2004, On page(s): 55 - 59
ISSN: 1582-7445, e-ISSN: 1844-7600
Digital Object Identifier: Not assigned
Dynamic computing environments require adaptive algorithms for load balancing. To increase system flexibility, mobile agents are used for encapsulating tasks. During successive migration phases, those agents have the possibility to explore, to learn and to share information about load. Ant Colony Optimisation metaheuristic is used for routing tasks with high efficiency due to its capacity to store and to learn from systems history. Finally, this information is used in maintaining an equitable task load for each processing unit. MPI implementations were used to make simulations and comparisons with different deterministic nearest neighbour load balancing techniques. Results have shown some improvements in terms of convergence stability and task distribution compared with classical algorithms.
On-line references are not available - see the PDF file if available.
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania
All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.
Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.
Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.