A SIMD interpreter for Genetic Programming on GPU
Graphics Cards
W. Langdon. CSM-470. Department of Computer Science, University of Essex, Colchester, UK, (3 July 2007)
Zusammenfassung
Mackey-Glass chaotic time series prediction and
non-nuclear protein classification show the feasibility
of evaluating genetic programming populations on SPMD
parallel computing consumer gaming graphics processing
units. The C++ framework with a regular disk less Linux
KDE desktop equipped with a single leading nVidia
GeForce 8800 GTX graphics processing unit card is
demonstrated evolving programs at Giga GP operation per
second (895 million GPops). The RapidMind general
processing on GPU (GPGPU) framework supports evaluating
an entire population of a quarter of a million
individual programs on a non-trivial problem in 4
seconds. An efficient reverse polish notation (RPN)
tree based GP is given.
%0 Report
%1 langdon:2007:csm470
%A Langdon, W. B.
%C Colchester, UK
%D 2007
%K gp
%N CSM-470
%T A SIMD interpreter for Genetic Programming on GPU
Graphics Cards
%U http://cswww.essex.ac.uk/technical-reports/2007/csm_470.pdf
%X Mackey-Glass chaotic time series prediction and
non-nuclear protein classification show the feasibility
of evaluating genetic programming populations on SPMD
parallel computing consumer gaming graphics processing
units. The C++ framework with a regular disk less Linux
KDE desktop equipped with a single leading nVidia
GeForce 8800 GTX graphics processing unit card is
demonstrated evolving programs at Giga GP operation per
second (895 million GPops). The RapidMind general
processing on GPU (GPGPU) framework supports evaluating
an entire population of a quarter of a million
individual programs on a non-trivial problem in 4
seconds. An efficient reverse polish notation (RPN)
tree based GP is given.
@techreport{langdon:2007:csm470,
abstract = {Mackey-Glass chaotic time series prediction and
non-nuclear protein classification show the feasibility
of evaluating genetic programming populations on SPMD
parallel computing consumer gaming graphics processing
units. The C++ framework with a regular disk less Linux
KDE desktop equipped with a single leading nVidia
GeForce 8800 GTX graphics processing unit card is
demonstrated evolving programs at Giga GP operation per
second (895 million GPops). The RapidMind general
processing on GPU (GPGPU) framework supports evaluating
an entire population of a quarter of a million
individual programs on a non-trivial problem in 4
seconds. An efficient reverse polish notation (RPN)
tree based GP is given.},
added-at = {2016-03-20T19:25:38.000+0100},
address = {Colchester, UK},
author = {Langdon, W. B.},
biburl = {https://www.bibsonomy.org/bibtex/2e28e8c741f70d0089c9e304ab3988797/hayral},
institution = {Department of Computer Science, University of Essex},
interhash = {3af8cf6cd5ffd16faca94bacc8ac4423},
intrahash = {e28e8c741f70d0089c9e304ab3988797},
issn = {1744-8050},
keywords = {gp},
month = {3 July},
notes = {Memorial University. Replaced by
\cite{langdon:2008:eurogp}},
number = {CSM-470},
size = {16 pages},
timestamp = {2016-03-20T19:25:38.000+0100},
title = {A {SIMD} interpreter for Genetic Programming on {GPU}
Graphics Cards},
url = {http://cswww.essex.ac.uk/technical-reports/2007/csm_470.pdf},
year = 2007
}