Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Genetic algorithms using SISAL parallel programming language

Conference ·
OSTI ID:105141

Genetic algorithms are a mathematical optimization technique developed by John Holland at the University of Michigan [1]. The SISAL programming language possesses many of the characteristics desired to implement genetic algorithms. SISAL is a deterministic, functional programming language which is inherently parallel. Because SISAL is functional and based on mathematical concepts, genetic algorithms can be efficiently translated into the language. Several of the steps involved in genetic algorithms, such as mutation, crossover, and fitness evaluation, can be parallelized using SISAL. In this paper I will l discuss the implementation and performance of parallel genetic algorithms in SISAL.

Research Organization:
Lawrence Livermore National Lab., CA (United States)
OSTI ID:
105141
Report Number(s):
UCRL-ID--114972-2; CONF-9405328--Absts.; ON: DE95014118
Country of Publication:
United States
Language:
English

Similar Records

Proceedings of the second SISAL users` conference
Conference · Mon Nov 30 23:00:00 EST 1992 · OSTI ID:10129865

SISAL: Toward resolving the parallel programming crisis
Conference · Sun Mar 01 23:00:00 EST 1992 · OSTI ID:5441572

SISAL: Toward resolving the parallel programming crisis
Conference · Sun Mar 01 23:00:00 EST 1992 · OSTI ID:10137020