A comparison of OpenMP and MPI for neural network simulations on a SunFire 6800

被引：0

作者：

Strey, A ^{[1
]}

机构：

[1] Univ Ulm, Dept Neural Informat Proc, D-89069 Ulm, Germany

来源：

PARALLEL COMPUTING: SOFTWARE TECHNOLOGY, ALGORITHMS, ARCHITECTURES AND APPLICATIONS | 2004年 / 13卷

关键词：

D O I：

暂无

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

This paper discusses several possibilities for the parallel implementation of a two-layer artificial neural network on a Symmetric Multiprocessor (SMP). Thread-parallel implementations based on OpenMP and process-parallel implementations based on the MPI communication library are compared. Different data and work partitioning strategies are investigated and the performance of all implementations is evaluated on a SunFire 6800.

引用

页码：201 / 208

页数：8

共 9 条

[1]

BANE M, 2000, P CRAY US GROUP SUMM

[2]

Bull J.M., 1999, P 1 EUR WORKSH OPENM, P99

[3]

CHARLESWORTH AE, 2001, P SUPERCOMPUTING

[4] SPEEDING UP BACKPROPAGATION TRAINING ON A HYPERCUBE COMPUTER [J].

KERCKHOFFS, EJH ;

WEDMAN, FW ;

FRIETMAN, EEE .

NEUROCOMPUTING, 1992, 4 (1-2) :43-63

[5]

Krawezik G., 2003, P 15 ANN ACM S PAR A, P118

[6] NETWORKS FOR APPROXIMATION AND LEARNING [J].

POGGIO, T ;

GIROSI, F .

PROCEEDINGS OF THE IEEE, 1990, 78 (09) :1481-1497

[7]

POMERLEAU DA, 1988, P IEEE INT C NEURAL, P143

[8] Three learning phases for radial-basis-function networks [J].

Schwenker, F ;

Kestler, HA ;

Palm, G .

NEURAL NETWORKS, 2001, 14 (4-5) :439-458

[9] Simulating artificial neural networks on parallel architectures [J].

Serbedzija, NB .

COMPUTER, 1996, 29 (03) :56-&

← 1 →