Toward exploitation of cell multi-processor array in time-consuming applications by using CNN model

Zoltán Nagy, László Kék, Zoltán Kincses, András Kiss, P. Szolgay

Research output: Article

7 Citations (Scopus)

Abstract

Array computers can be useful in the solution of numerical spatio-temporal problems such as the state equation of the CNN or partial differential equations (PDE). IBM has recently introduced the cell broadband engine (cell BE) architecture, which contains eight identical vector processors in an array structure. In this paper the implementation of CNN simulation kernel on the cell BE is described. The simulation kernel is optimized according to the special requirements of the cell BE and can implement linear as well as nonlinear (piecewise linear) templates. In addition, the 2-D Princeton ocean model is implemented on the cell BE. The area/speed/power tradeoffs of our solution and different hardware implementations are also compared.

Original languageEnglish
Pages (from-to)605-622
Number of pages18
JournalInternational Journal of Circuit Theory and Applications
Volume36
Issue number5-6
DOIs
Publication statusPublished - júl. 2008

Fingerprint

Cell Broadband Engine
Parallel processing systems
Multiprocessor
Exploitation
Engines
Cell
kernel
Hardware Implementation
Equations of state
Equation of State
Piecewise Linear
Model
Ocean
Partial differential equations
Template
Simulation
Partial differential equation
Trade-offs
Hardware
Requirements

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Toward exploitation of cell multi-processor array in time-consuming applications by using CNN model. / Nagy, Zoltán; Kék, László; Kincses, Zoltán; Kiss, András; Szolgay, P.

In: International Journal of Circuit Theory and Applications, Vol. 36, No. 5-6, 07.2008, p. 605-622.

Research output: Article

Nagy, Zoltán ; Kék, László ; Kincses, Zoltán ; Kiss, András ; Szolgay, P. / Toward exploitation of cell multi-processor array in time-consuming applications by using CNN model. In: International Journal of Circuit Theory and Applications. 2008 ; Vol. 36, No. 5-6. pp. 605-622.
@article{46ac24b45b7149229056f5fe8cee7adb,
title = "Toward exploitation of cell multi-processor array in time-consuming applications by using CNN model",
abstract = "Array computers can be useful in the solution of numerical spatio-temporal problems such as the state equation of the CNN or partial differential equations (PDE). IBM has recently introduced the cell broadband engine (cell BE) architecture, which contains eight identical vector processors in an array structure. In this paper the implementation of CNN simulation kernel on the cell BE is described. The simulation kernel is optimized according to the special requirements of the cell BE and can implement linear as well as nonlinear (piecewise linear) templates. In addition, the 2-D Princeton ocean model is implemented on the cell BE. The area/speed/power tradeoffs of our solution and different hardware implementations are also compared.",
keywords = "Cellular neural networks, Emulated digital CNN-UM, Nonlinear CNN template partial differential equations, Ocean modeling",
author = "Zolt{\'a}n Nagy and L{\'a}szl{\'o} K{\'e}k and Zolt{\'a}n Kincses and Andr{\'a}s Kiss and P. Szolgay",
year = "2008",
month = "7",
doi = "10.1002/cta.508",
language = "English",
volume = "36",
pages = "605--622",
journal = "International Journal of Circuit Theory and Applications",
issn = "0098-9886",
publisher = "John Wiley and Sons Ltd",
number = "5-6",

}

TY - JOUR

T1 - Toward exploitation of cell multi-processor array in time-consuming applications by using CNN model

AU - Nagy, Zoltán

AU - Kék, László

AU - Kincses, Zoltán

AU - Kiss, András

AU - Szolgay, P.

PY - 2008/7

Y1 - 2008/7

N2 - Array computers can be useful in the solution of numerical spatio-temporal problems such as the state equation of the CNN or partial differential equations (PDE). IBM has recently introduced the cell broadband engine (cell BE) architecture, which contains eight identical vector processors in an array structure. In this paper the implementation of CNN simulation kernel on the cell BE is described. The simulation kernel is optimized according to the special requirements of the cell BE and can implement linear as well as nonlinear (piecewise linear) templates. In addition, the 2-D Princeton ocean model is implemented on the cell BE. The area/speed/power tradeoffs of our solution and different hardware implementations are also compared.

AB - Array computers can be useful in the solution of numerical spatio-temporal problems such as the state equation of the CNN or partial differential equations (PDE). IBM has recently introduced the cell broadband engine (cell BE) architecture, which contains eight identical vector processors in an array structure. In this paper the implementation of CNN simulation kernel on the cell BE is described. The simulation kernel is optimized according to the special requirements of the cell BE and can implement linear as well as nonlinear (piecewise linear) templates. In addition, the 2-D Princeton ocean model is implemented on the cell BE. The area/speed/power tradeoffs of our solution and different hardware implementations are also compared.

KW - Cellular neural networks

KW - Emulated digital CNN-UM

KW - Nonlinear CNN template partial differential equations

KW - Ocean modeling

UR - http://www.scopus.com/inward/record.url?scp=47749086315&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47749086315&partnerID=8YFLogxK

U2 - 10.1002/cta.508

DO - 10.1002/cta.508

M3 - Article

AN - SCOPUS:47749086315

VL - 36

SP - 605

EP - 622

JO - International Journal of Circuit Theory and Applications

JF - International Journal of Circuit Theory and Applications

SN - 0098-9886

IS - 5-6

ER -