A stochastic genetic algorithm for dynamic load balancing in distributed systems

Author

Munetomo, Masaharu ; Takai, Yoshiaki ; Sato, Yoshiharu

Author_Institution

Dept. of Inf. & Data Anal., Hokkaido Univ., Sapporo, Japan

Volume

4

fYear

1995

fDate

22-25 Oct 1995

Firstpage

3795

Abstract

This paper presents a genetic algorithm (GA) for stochastic environments and its application to dynamic load balancing in distributed systems. We have proposed a stochastic genetic algorithm (StGA) which has an evaluation mechanism for fitness values based on the reinforcement learning in order to adapt to stochastic environments. We apply the StGA to the decision phase of task migration requests in dynamic load balancing, and we realize a task distribution system based on the StGA in a local area network which consists of UNIX workstations

Keywords

distributed processing; genetic algorithms; learning (artificial intelligence); resource allocation; stochastic processes; LAN; UNIX workstations; distributed systems; dynamic load balancing; evaluation mechanism; fitness values; local area network; reinforcement learning; stochastic genetic algorithm; task distribution system; task migration requests; Convergence; Data analysis; Genetic algorithms; Learning automata; Load management; Local area networks; Stochastic processes; Stochastic systems; Systems engineering and theory; Workstations;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man and Cybernetics, 1995. Intelligent Systems for the 21st Century., IEEE International Conference on

Conference_Location

Vancouver, BC

Print_ISBN

0-7803-2559-1

Type

conf

DOI

10.1109/ICSMC.1995.538379

Filename

538379