Scalable architecture for parallel distributed implementation of genetic programming on network of workstations
Created by W.Langdon from
gp-bibliography.bib Revision:1.7954
- @Article{tanev:2001:SA,
-
author = "Ivan Tanev and Takashi Uozumi and Koichi Ono",
-
title = "Scalable architecture for parallel distributed
implementation of genetic programming on network of
workstations",
-
journal = "Journal of Systems Architecture",
-
volume = "47",
-
pages = "557--572",
-
year = "2001",
-
number = "7",
-
month = jul,
-
keywords = "genetic algorithms, genetic programming, Distributed
component object model, Island model of parallelism,
Network of workstations",
-
ISSN = "1383-7621",
-
DOI = "doi:10.1016/S1383-7621(01)00015-7",
-
URL = "http://www.sciencedirect.com/science/article/B6V1F-43RV156-2/2/96f14334f4a466a6a7a6034c398ff8c4",
-
size = "16 pages",
-
abstract = "We present an approach for developing a scalable
architecture for parallel distributed implementation of
genetic programming (PDIGP). The approach is based on
exploitation of the inherent parallelism among
semi-isolated subpopulations in genetic programming
(GP). Proposed implementation runs on cost-efficient
configurations of networks on workstations in LAN and
Internet environment. Developed architecture features
single global migration broker and centralized manager
of the semi-isolated subpopulations, which contribute
to achieving quick propagation of the globally fittest
individuals among the subpopulations, reducing the
performance demands to the communication network, and
achieving flexibility in system configurations by
introducing dynamically scaling up opportunities. PDIGP
exploits distributed component object model (DCOM) as a
communication paradigm, which as a true system model
offers generic support for the issues of naming,
locating and protecting the distributed entities in
proposed architecture of PDIGP. Experimentally obtained
results of computational effort of proposed PDIGP are
discussed. The results show that computational effort
of PDIGP marginally differs from the computational
effort in canonical panmictic GP evolving single large
population. For PDIGP running on systems configurations
with 16 workstations the computational effort is less
than panmictic GP, while for smaller configurations it
is insignificantly more. Analytically obtained and
empirically proved results of the speedup of
computational performance indicate that PDIGP features
linear, close to ideal characteristics. Experimentally
obtained results of PDIGP running on configurations
with eight workstations show close to 8-fold overall
speedup. These results are consistent with the
anticipated cumulative effect of the insignificant
increase of computational effort for the considered
configuration and the close to linear speedup of
computational performance.",
- }
Genetic Programming entries for
Ivan T Tanev
Takashi Uozumi
Koichi Ono
Citations