Implementation and evaluation of self-organizing map algorithm on a graphic processor

Akihiro Shitara; Yuri Nishikawa; Masato Yoshimi; Hideharu Amano

Implementation and evaluation of self-organizing map algorithm on a graphic processor

Akihiro Shitara, Yuri Nishikawa, Masato Yoshimi, Hideharu Amano

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper, we introduce an implementation of algorithm for self-organizing maps(SOM) using GPUs and discuss its evaluation. We used CUDA provided by NVIDIA Corporation for parallel programming, profiling, and data flow optimization so as to exploit inherent data-level parallelism of the algorithm. By using three NVIDIA's graphic cards for evaluation, we investigated the relationships among the number of processor elements, amount of memory device and performance. As the result of performance evaluation with various parameter combinations, we found that implementation on GTX280 achieved 150 times higher performance of Intel Core 2 Quad 2.40 GHz when parameters of map size, dimension of vectors and learning size were 1372×1372, 128 and 128, respectively.

Original language	English
Title of host publication	Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009
Pages	253-260
Number of pages	8
Publication status	Published - 2009 Dec 1
Externally published	Yes
Event	21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009 - Cambridge, MA, United States Duration: 2009 Nov 2 → 2009 Nov 4

Publication series

Name	Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems
ISSN (Print)	1027-2658

Other

Other	21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009
Country/Territory	United States
City	Cambridge, MA
Period	09/11/2 → 09/11/4

Keywords

CUDA
GPGPU
GPU
Self-organizing Map

ASJC Scopus subject areas

Software
Hardware and Architecture
Computer Networks and Communications

Cite this

Shitara, A., Nishikawa, Y., Yoshimi, M., & Amano, H. (2009). Implementation and evaluation of self-organizing map algorithm on a graphic processor. In Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009 (pp. 253-260). (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems).

Implementation and evaluation of self-organizing map algorithm on a graphic processor. / Shitara, Akihiro; Nishikawa, Yuri; Yoshimi, Masato et al.
Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009. 2009. p. 253-260 (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Shitara, A, Nishikawa, Y, Yoshimi, M & Amano, H 2009, Implementation and evaluation of self-organizing map algorithm on a graphic processor. in Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009. Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems, pp. 253-260, 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009, Cambridge, MA, United States, 09/11/2.

Shitara A, Nishikawa Y, Yoshimi M, Amano H. Implementation and evaluation of self-organizing map algorithm on a graphic processor. In Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009. 2009. p. 253-260. (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems).

Shitara, Akihiro ; Nishikawa, Yuri ; Yoshimi, Masato et al. / Implementation and evaluation of self-organizing map algorithm on a graphic processor. Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009. 2009. pp. 253-260 (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems).

@inproceedings{69cbeb9fb1b748c3b2b7b1cec2270573,

title = "Implementation and evaluation of self-organizing map algorithm on a graphic processor",

abstract = "In this paper, we introduce an implementation of algorithm for self-organizing maps(SOM) using GPUs and discuss its evaluation. We used CUDA provided by NVIDIA Corporation for parallel programming, profiling, and data flow optimization so as to exploit inherent data-level parallelism of the algorithm. By using three NVIDIA's graphic cards for evaluation, we investigated the relationships among the number of processor elements, amount of memory device and performance. As the result of performance evaluation with various parameter combinations, we found that implementation on GTX280 achieved 150 times higher performance of Intel Core 2 Quad 2.40 GHz when parameters of map size, dimension of vectors and learning size were 1372×1372, 128 and 128, respectively.",

keywords = "CUDA, GPGPU, GPU, Self-organizing Map",

author = "Akihiro Shitara and Yuri Nishikawa and Masato Yoshimi and Hideharu Amano",

year = "2009",

month = dec,

day = "1",

language = "English",

isbn = "9780889868113",

series = "Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems",

pages = "253--260",

booktitle = "Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009",

note = "21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009 ; Conference date: 02-11-2009 Through 04-11-2009",

}

TY - GEN

T1 - Implementation and evaluation of self-organizing map algorithm on a graphic processor

AU - Shitara, Akihiro

AU - Nishikawa, Yuri

AU - Yoshimi, Masato

AU - Amano, Hideharu

PY - 2009/12/1

Y1 - 2009/12/1

N2 - In this paper, we introduce an implementation of algorithm for self-organizing maps(SOM) using GPUs and discuss its evaluation. We used CUDA provided by NVIDIA Corporation for parallel programming, profiling, and data flow optimization so as to exploit inherent data-level parallelism of the algorithm. By using three NVIDIA's graphic cards for evaluation, we investigated the relationships among the number of processor elements, amount of memory device and performance. As the result of performance evaluation with various parameter combinations, we found that implementation on GTX280 achieved 150 times higher performance of Intel Core 2 Quad 2.40 GHz when parameters of map size, dimension of vectors and learning size were 1372×1372, 128 and 128, respectively.

AB - In this paper, we introduce an implementation of algorithm for self-organizing maps(SOM) using GPUs and discuss its evaluation. We used CUDA provided by NVIDIA Corporation for parallel programming, profiling, and data flow optimization so as to exploit inherent data-level parallelism of the algorithm. By using three NVIDIA's graphic cards for evaluation, we investigated the relationships among the number of processor elements, amount of memory device and performance. As the result of performance evaluation with various parameter combinations, we found that implementation on GTX280 achieved 150 times higher performance of Intel Core 2 Quad 2.40 GHz when parameters of map size, dimension of vectors and learning size were 1372×1372, 128 and 128, respectively.

KW - CUDA

KW - GPGPU

KW - GPU

KW - Self-organizing Map

UR - http://www.scopus.com/inward/record.url?scp=77952386820&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952386820&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:77952386820

SN - 9780889868113

T3 - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems

SP - 253

EP - 260

BT - Proceedings of the 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009

T2 - 21st IASTED International Conference on Parallel and Distributed Computing and Systems, PDCS 2009

Y2 - 2 November 2009 through 4 November 2009

ER -

Implementation and evaluation of self-organizing map algorithm on a graphic processor

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this