Parallel execution of SVM training using graphics processing units (SVMTrGPUs)

Nur Shakirah Md Salleh, Muhammad Fahim Baharim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medical diagnosis, national defense, and web search engine. This paper focuses on the implementation of pattern classification technique, Support Vector Machine (SVM) using vector processor approach. We have carried out a performance analysis to benchmark the sequential SVM program against the Graphics Processing Units (GPUs) optimization. The result shows that the parallelization of SVM training duration achieves a better performance than the sequential code speedups by 6.40.

Original languageEnglish
Title of host publicationProceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages260-263
Number of pages4
ISBN (Electronic)9781479982523
DOIs
Publication statusPublished - 31 May 2016
Event5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015 - Batu Ferringhi, Penang, Malaysia
Duration: 27 Nov 201529 Nov 2015

Other

Other5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015
CountryMalaysia
CityBatu Ferringhi, Penang
Period27/11/1529/11/15

Fingerprint

Support vector machines
Pattern recognition
Parallel processing systems
Search engines
Graphics processing unit

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Artificial Intelligence

Cite this

Md Salleh, N. S., & Baharim, M. F. (2016). Parallel execution of SVM training using graphics processing units (SVMTrGPUs). In Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015 (pp. 260-263). [7482194] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCSCE.2015.7482194
Md Salleh, Nur Shakirah ; Baharim, Muhammad Fahim. / Parallel execution of SVM training using graphics processing units (SVMTrGPUs). Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 260-263
@inproceedings{513e289e4b3445619b422ef552eb93ce,
title = "Parallel execution of SVM training using graphics processing units (SVMTrGPUs)",
abstract = "Parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medical diagnosis, national defense, and web search engine. This paper focuses on the implementation of pattern classification technique, Support Vector Machine (SVM) using vector processor approach. We have carried out a performance analysis to benchmark the sequential SVM program against the Graphics Processing Units (GPUs) optimization. The result shows that the parallelization of SVM training duration achieves a better performance than the sequential code speedups by 6.40.",
author = "{Md Salleh}, {Nur Shakirah} and Baharim, {Muhammad Fahim}",
year = "2016",
month = "5",
day = "31",
doi = "10.1109/ICCSCE.2015.7482194",
language = "English",
pages = "260--263",
booktitle = "Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Md Salleh, NS & Baharim, MF 2016, Parallel execution of SVM training using graphics processing units (SVMTrGPUs). in Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015., 7482194, Institute of Electrical and Electronics Engineers Inc., pp. 260-263, 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015, Batu Ferringhi, Penang, Malaysia, 27/11/15. https://doi.org/10.1109/ICCSCE.2015.7482194

Parallel execution of SVM training using graphics processing units (SVMTrGPUs). / Md Salleh, Nur Shakirah; Baharim, Muhammad Fahim.

Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 260-263 7482194.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Parallel execution of SVM training using graphics processing units (SVMTrGPUs)

AU - Md Salleh, Nur Shakirah

AU - Baharim, Muhammad Fahim

PY - 2016/5/31

Y1 - 2016/5/31

N2 - Parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medical diagnosis, national defense, and web search engine. This paper focuses on the implementation of pattern classification technique, Support Vector Machine (SVM) using vector processor approach. We have carried out a performance analysis to benchmark the sequential SVM program against the Graphics Processing Units (GPUs) optimization. The result shows that the parallelization of SVM training duration achieves a better performance than the sequential code speedups by 6.40.

AB - Parallel computing is a simultaneous use of multiple compute resources, for example, processors to solve complex computational problems. It has been used in high-end computing areas such as pattern recognition, medical diagnosis, national defense, and web search engine. This paper focuses on the implementation of pattern classification technique, Support Vector Machine (SVM) using vector processor approach. We have carried out a performance analysis to benchmark the sequential SVM program against the Graphics Processing Units (GPUs) optimization. The result shows that the parallelization of SVM training duration achieves a better performance than the sequential code speedups by 6.40.

UR - http://www.scopus.com/inward/record.url?scp=84978834001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978834001&partnerID=8YFLogxK

U2 - 10.1109/ICCSCE.2015.7482194

DO - 10.1109/ICCSCE.2015.7482194

M3 - Conference contribution

SP - 260

EP - 263

BT - Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Md Salleh NS, Baharim MF. Parallel execution of SVM training using graphics processing units (SVMTrGPUs). In Proceedings - 5th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 260-263. 7482194 https://doi.org/10.1109/ICCSCE.2015.7482194