A neural network approach for dynamic load balancing in homogeneous distributed systems

Author

El-Abd, Aly E. ; El-Bendary, Mohamed I.

Author_Institution

Dept. of Electron. & Comput. Eng., Acad. for Sci. & Technol., Alexandria, Egypt

Volume

1

fYear

1997

fDate

7-10 Jan 1997

Firstpage

628

Abstract

A novel neural-based solution to the problem of dynamic load balancing in homogeneous distributed systems is proposed. The winner-take-all (WTA) neural network model is used for implementing the selection and location policies of a typical dynamic load balancing algorithm. Tasks with interprocess communication requirements are considered. All delays due to any usage of the communications network resource are taken into account. Performance studied by simulation is compared against the no load balancing case. The results suggest that it is always beneficial to use load balancing than not at all

Keywords

delays; distributed algorithms; distributed memory systems; neural nets; resource allocation; virtual machines; communications network resource usage; delays; dynamic load balancing; dynamic load balancing algorithm; homogeneous distributed systems; interprocess communication requirements; location policies; performance; selection policies; simulation; winner-take-all neural network model; Communication networks; Costs; Distributed computing; Heuristic algorithms; Intelligent networks; Load management; Message passing; Neural networks; Neurons; Power system modeling; System performance;

fLanguage

English

Publisher

ieee

Conference_Titel

System Sciences, 1997, Proceedings of the Thirtieth Hawaii International Conference on

Conference_Location

Wailea, HI

ISSN

1060-3425

Print_ISBN

0-8186-7743-0

Type

conf

DOI

10.1109/HICSS.1997.667421

Filename

667421