Evaluation of task clustering algorithm by FFT for heterogeneous distributed system

Shuya Hashimoto, Emilia Ndilokelwa Weyulu, Kazuo Hajikano, Hidehiro Kanemitsu, Moo Wan Kim

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

1 Citation (Scopus)

Abstract

In this paper, the evaluation result of a task clustering heuristic algorithm proposed for heterogeneous distributed systems is shown. The proposed heuristic algorithm is based on our original concept, known as 'Worst Schedule Length (WSL)'. It derives the lower bound of the total execution time of the cluster for each processor using WSL, then the processor which contributes to minimize WSL is chosen as an assignment target. Task clustering is then performed to get minimal response time (i.e., minimal schedule length). We show that our proposed method has advantages over existing conventional approaches through the evaluation results.

Original languageEnglish
Title of host publication19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages94-97
Number of pages4
ISBN (Electronic)9788996865094
DOIs
Publication statusPublished - 2017 Mar 29
Event19th International Conference on Advanced Communications Technology, ICACT 2017 - Pyeongchang, Korea, Republic of
Duration: 2017 Feb 192017 Feb 22

Other

Other19th International Conference on Advanced Communications Technology, ICACT 2017
CountryKorea, Republic of
CityPyeongchang
Period17/2/1917/2/22

Fingerprint

Heuristic algorithms
Clustering algorithms
Fast Fourier transforms

Keywords

  • Big data
  • Distributed processing
  • Fast fourier transform
  • Task graph

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Hashimoto, S., Weyulu, E. N., Hajikano, K., Kanemitsu, H., & Kim, M. W. (2017). Evaluation of task clustering algorithm by FFT for heterogeneous distributed system. In 19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding (pp. 94-97). [7890064] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/ICACT.2017.7890064

Evaluation of task clustering algorithm by FFT for heterogeneous distributed system. / Hashimoto, Shuya; Weyulu, Emilia Ndilokelwa; Hajikano, Kazuo; Kanemitsu, Hidehiro; Kim, Moo Wan.

19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding. Institute of Electrical and Electronics Engineers Inc., 2017. p. 94-97 7890064.

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

Hashimoto, S, Weyulu, EN, Hajikano, K, Kanemitsu, H & Kim, MW 2017, Evaluation of task clustering algorithm by FFT for heterogeneous distributed system. in 19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding., 7890064, Institute of Electrical and Electronics Engineers Inc., pp. 94-97, 19th International Conference on Advanced Communications Technology, ICACT 2017, Pyeongchang, Korea, Republic of, 17/2/19. https://doi.org/10.23919/ICACT.2017.7890064
Hashimoto S, Weyulu EN, Hajikano K, Kanemitsu H, Kim MW. Evaluation of task clustering algorithm by FFT for heterogeneous distributed system. In 19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding. Institute of Electrical and Electronics Engineers Inc. 2017. p. 94-97. 7890064 https://doi.org/10.23919/ICACT.2017.7890064
Hashimoto, Shuya ; Weyulu, Emilia Ndilokelwa ; Hajikano, Kazuo ; Kanemitsu, Hidehiro ; Kim, Moo Wan. / Evaluation of task clustering algorithm by FFT for heterogeneous distributed system. 19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 94-97
@inproceedings{c0482a186ded40a884933babab93321c,
title = "Evaluation of task clustering algorithm by FFT for heterogeneous distributed system",
abstract = "In this paper, the evaluation result of a task clustering heuristic algorithm proposed for heterogeneous distributed systems is shown. The proposed heuristic algorithm is based on our original concept, known as 'Worst Schedule Length (WSL)'. It derives the lower bound of the total execution time of the cluster for each processor using WSL, then the processor which contributes to minimize WSL is chosen as an assignment target. Task clustering is then performed to get minimal response time (i.e., minimal schedule length). We show that our proposed method has advantages over existing conventional approaches through the evaluation results.",
keywords = "Big data, Distributed processing, Fast fourier transform, Task graph",
author = "Shuya Hashimoto and Weyulu, {Emilia Ndilokelwa} and Kazuo Hajikano and Hidehiro Kanemitsu and Kim, {Moo Wan}",
year = "2017",
month = "3",
day = "29",
doi = "10.23919/ICACT.2017.7890064",
language = "English",
pages = "94--97",
booktitle = "19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Evaluation of task clustering algorithm by FFT for heterogeneous distributed system

AU - Hashimoto, Shuya

AU - Weyulu, Emilia Ndilokelwa

AU - Hajikano, Kazuo

AU - Kanemitsu, Hidehiro

AU - Kim, Moo Wan

PY - 2017/3/29

Y1 - 2017/3/29

N2 - In this paper, the evaluation result of a task clustering heuristic algorithm proposed for heterogeneous distributed systems is shown. The proposed heuristic algorithm is based on our original concept, known as 'Worst Schedule Length (WSL)'. It derives the lower bound of the total execution time of the cluster for each processor using WSL, then the processor which contributes to minimize WSL is chosen as an assignment target. Task clustering is then performed to get minimal response time (i.e., minimal schedule length). We show that our proposed method has advantages over existing conventional approaches through the evaluation results.

AB - In this paper, the evaluation result of a task clustering heuristic algorithm proposed for heterogeneous distributed systems is shown. The proposed heuristic algorithm is based on our original concept, known as 'Worst Schedule Length (WSL)'. It derives the lower bound of the total execution time of the cluster for each processor using WSL, then the processor which contributes to minimize WSL is chosen as an assignment target. Task clustering is then performed to get minimal response time (i.e., minimal schedule length). We show that our proposed method has advantages over existing conventional approaches through the evaluation results.

KW - Big data

KW - Distributed processing

KW - Fast fourier transform

KW - Task graph

UR - http://www.scopus.com/inward/record.url?scp=85018526868&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018526868&partnerID=8YFLogxK

U2 - 10.23919/ICACT.2017.7890064

DO - 10.23919/ICACT.2017.7890064

M3 - Conference contribution

AN - SCOPUS:85018526868

SP - 94

EP - 97

BT - 19th International Conference on Advanced Communications Technology: Opening Era of Smart Society, ICACT 2017 - Proceeding

PB - Institute of Electrical and Electronics Engineers Inc.

ER -