Parallel backpropagation learning algorithms on CRAY Y-MP8/864 supercomputer

Shih-Lin Hung; H. Adeli

doi:10.1016/0925-2312(93)90042-2

Parallel backpropagation learning algorithms on CRAY Y-MP8/864 supercomputer

Shih-Lin Hung, H. Adeli^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

110 Scopus citations

Abstract

Parallel backpropagation neural networks learning algorithms have been developed employing the vectorization and microtasking capabilities of vector MIMD machines. They have been implemented in C on CRAY Y-MP/864 supercomputer under UNICOS operating system. The algorithms have been applied to two different domains: engineering design and image recognition, and their performance has been investigated. A maximum speedup of about 6.7 is achieved using eight processors for a large network with 5950 links due to microtasking only. When vectorization is combined with microtasking, a maximum speedup of about 33 is realized using eight processors.

Original language	English
Pages (from-to)	287-302
Number of pages	16
Journal	Neurocomputing
Volume	5
Issue number	6
DOIs	https://doi.org/10.1016/0925-2312(93)90042-2
State	Published - 1 Jan 1993

Keywords

Backpropagation
multitasking
neural networks
parallel processing
vectorization

Access to Document

10.1016/0925-2312(93)90042-2

Cite this

@article{1145c8eee38b425ea0e2afcd936b93ed,

title = "Parallel backpropagation learning algorithms on CRAY Y-MP8/864 supercomputer",

abstract = "Parallel backpropagation neural networks learning algorithms have been developed employing the vectorization and microtasking capabilities of vector MIMD machines. They have been implemented in C on CRAY Y-MP/864 supercomputer under UNICOS operating system. The algorithms have been applied to two different domains: engineering design and image recognition, and their performance has been investigated. A maximum speedup of about 6.7 is achieved using eight processors for a large network with 5950 links due to microtasking only. When vectorization is combined with microtasking, a maximum speedup of about 33 is realized using eight processors.",

keywords = "Backpropagation, multitasking, neural networks, parallel processing, vectorization",

author = "Shih-Lin Hung and H. Adeli",

year = "1993",

month = jan,

day = "1",

doi = "10.1016/0925-2312(93)90042-2",

language = "English",

volume = "5",

pages = "287--302",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

number = "6",

}

TY - JOUR

T1 - Parallel backpropagation learning algorithms on CRAY Y-MP8/864 supercomputer

AU - Hung, Shih-Lin

AU - Adeli, H.

PY - 1993/1/1

Y1 - 1993/1/1

N2 - Parallel backpropagation neural networks learning algorithms have been developed employing the vectorization and microtasking capabilities of vector MIMD machines. They have been implemented in C on CRAY Y-MP/864 supercomputer under UNICOS operating system. The algorithms have been applied to two different domains: engineering design and image recognition, and their performance has been investigated. A maximum speedup of about 6.7 is achieved using eight processors for a large network with 5950 links due to microtasking only. When vectorization is combined with microtasking, a maximum speedup of about 33 is realized using eight processors.

AB - Parallel backpropagation neural networks learning algorithms have been developed employing the vectorization and microtasking capabilities of vector MIMD machines. They have been implemented in C on CRAY Y-MP/864 supercomputer under UNICOS operating system. The algorithms have been applied to two different domains: engineering design and image recognition, and their performance has been investigated. A maximum speedup of about 6.7 is achieved using eight processors for a large network with 5950 links due to microtasking only. When vectorization is combined with microtasking, a maximum speedup of about 33 is realized using eight processors.

KW - Backpropagation

KW - multitasking

KW - neural networks

KW - parallel processing

KW - vectorization

UR - http://www.scopus.com/inward/record.url?scp=0027364993&partnerID=8YFLogxK

U2 - 10.1016/0925-2312(93)90042-2

DO - 10.1016/0925-2312(93)90042-2

M3 - Article

AN - SCOPUS:0027364993

SN - 0925-2312

VL - 5

SP - 287

EP - 302

JO - Neurocomputing

JF - Neurocomputing

IS - 6

ER -

Parallel backpropagation learning algorithms on CRAY Y-MP8/864 supercomputer

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this