Přístupnostní navigace
E-application
Search Search Close
Publication detail
JAROŠ, J.
Original Title
Multi-GPU Island-Based Genetic Algorithm Solving the Knapsack Problem
Type
conference paper
Language
English
Original Abstract
This paper introduces a novel implementation of the genetic algorithm exploiting a multi-GPU cluster. The proposed implementation employs an island-based genetic algorithm where every GPU evolves a single island. The individuals are treated by CUDA warps enabling to solve large knapsack instances and eliminate undesirable thread divergence. The OpenMPI interface is used to exchange genetic material among different islands and to collect statistical data. The implementation characteristics are investigated on a cluster composed of 14 GPUs and 4 six-core Intel Xeon processors.
Keywords
GPU; CUDA; MPI; GA; island model; knapsack
Authors
RIV year
2012
Released
11. 6. 2012
Publisher
Institute of Electrical and Electronics Engineers
Location
CA
ISBN
978-1-4673-1508-1
Book
2012 IEEE Congress on Evolutionary Computation, CEC 2012
Pages from
217
Pages to
224
Pages count
8
URL
https://www.fit.vut.cz/research/publication/9860/
BibTex
@inproceedings{BUT96925, author="Jiří {Jaroš}", title="Multi-GPU Island-Based Genetic Algorithm Solving the Knapsack Problem", booktitle="2012 IEEE Congress on Evolutionary Computation, CEC 2012", year="2012", pages="217--224", publisher="Institute of Electrical and Electronics Engineers", address="CA", doi="10.1109/CEC.2012.6256131", isbn="978-1-4673-1508-1", url="https://www.fit.vut.cz/research/publication/9860/" }