
Mathematical Foundations of Informatics and Programming
On mapping graphs of parallel programs onto graphs of distributed computer systems by recurrent neural networks
M. S. Tarkov^{} ^{} A. V. Rzhanov Institute of Semiconductor Physics of SB RAS, Novosibirsk, Russia
Abstract:
A problem of mapping graphs of parallel programs onto graphs of distributed computer systems by recurrent neural networks is formulated. The network parameters providing the absence of incorrect solutions are experimentally determined. By introduction of a penalty coefficient into the Lyapunov function for the program graph edges noncoincided with the edges of the computer system, the optimal solutions are computed for mapping the “line” program graph onto a twodimensional torus. To increase the optimal solution probability a method of the mapping decomposition is proposed. The method essence is a reduction of the solution matrix to a blockdiagonal shape. For exclusion of incorrect solutions in mapping the line onto threedimensional torus, a recurrent Wang network is used because it is converged more rapidly than the Hopfield network.
Keywords:
mapping, graphs of parallel programs, distributed computer systems, Hopfield network, recurrent Wang network.
Full text:
PDF file (616 kB)
References:
PDF file
HTML file
UDC:
004.032.26(06)
Citation:
M. S. Tarkov, “On mapping graphs of parallel programs onto graphs of distributed computer systems by recurrent neural networks”, Prikl. Diskr. Mat., 2010, no. 4(10), 33–40
Citation in format AMSBIB
\Bibitem{Tar10}
\by M.~S.~Tarkov
\paper On mapping graphs of parallel programs onto graphs of distributed computer systems by recurrent neural networks
\jour Prikl. Diskr. Mat.
\yr 2010
\issue 4(10)
\pages 3340
\mathnet{http://mi.mathnet.ru/pdm258}
Linking options:
http://mi.mathnet.ru/eng/pdm258 http://mi.mathnet.ru/eng/pdm/y2010/i4/p33
Citing articles on Google Scholar:
Russian citations,
English citations
Related articles on Google Scholar:
Russian articles,
English articles

Number of views: 
This page:  93  Full text:  32  References:  8  First page:  1 
