Publication detail

A Totally Distributed Genetic Algorithm: From a Cellular System to the Mesh of Processors

SEKANINA, L., DVOŘÁK, V.

Original Title

A Totally Distributed Genetic Algorithm: From a Cellular System to the Mesh of Processors

English Title

A Totally Distributed Genetic Algorithm: From a Cellular System to the Mesh of Processors

Type

conference paper

Language

en

Original Abstract

The paper deals with properties of the totally distributed genetic algorithm RGA, initially designed for the PIG cellular system. We have adopted the algorithm to a mesh of processors. Simulations were performed using Transim tool in order to investigate performance of this new RGA algorithm independently of a given application. This way, characteristics such as efficiency, speedup, communication delays, the influence of chromosome length and fitness calculations are easily evaluated beforehand.

English abstract

The paper deals with properties of the totally distributed genetic algorithm RGA, initially designed for the PIG cellular system. We have adopted the algorithm to a mesh of processors. Simulations were performed using Transim tool in order to investigate performance of this new RGA algorithm independently of a given application. This way, characteristics such as efficiency, speedup, communication delays, the influence of chromosome length and fitness calculations are easily evaluated beforehand.

Keywords

Cellular system, digital circuits, parallel genetic algorithm, mesh of processors, performance estimation.

Released

01.01.2001

Publisher

Faculty of Electrical Engineering

Location

Prague

ISBN

1-56555-225-3

Book

Modelling and Simulation 2001

Pages from

539

Pages to

543

Pages count

5

URL

BibTex


@inproceedings{BUT5432,
  author="Lukáš {Sekanina} and Václav {Dvořák}",
  title="A Totally Distributed Genetic Algorithm: From a Cellular System to the Mesh of Processors",
  annote="The paper deals with properties of the totally distributed genetic algorithm RGA, initially designed for the PIG cellular system. We have adopted the algorithm to a mesh of processors. Simulations were performed using Transim tool in order to investigate performance of this new RGA algorithm independently of a given application. This way, characteristics such as efficiency, speedup, communication delays, the influence of chromosome length and fitness calculations are easily evaluated beforehand.",
  address="Faculty of Electrical Engineering",
  booktitle="Modelling and Simulation 2001",
  chapter="5432",
  institution="Faculty of Electrical Engineering",
  year="2001",
  month="january",
  pages="539--543",
  publisher="Faculty of Electrical Engineering",
  type="conference paper"
}