Stable HPC cluster management scheme through performance evaluation

Jun Weon Yoon, Tae Yeong Hong, Chan Yeol Park, Heonchang Yu

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

1 Citation (Scopus)

Abstract

HPC is representative tools for performing large-scale scientific calculation both academia and industry. Tachyon is a high-performance parallel computing system which constructed based on SUN Blade X6275. It composed of 3,200 computing nodes and infra-facilities. Also, this machine works with various software stacks such as file system, archive manager, compiler, debugger, parallel tools, etc. In this paper, we handle the requirements and requisites to build and manage an HPC cluster environment. In addition, we analyzed the history of batch job which include information performed by scheduler. By doing so, we are able to gauge the needs and performance of the next system to be introduced.

Original languageEnglish
Title of host publicationLecture Notes in Electrical Engineering
PublisherSpringer Verlag
Pages1017-1023
Number of pages7
Volume330
ISBN (Print)9783662454015
DOIs
Publication statusPublished - 2015 Jan 1
Event6th FTRA International Conference on Computer Science and its Applications, CSA 2014 - Guam, United States
Duration: 2014 Dec 172014 Dec 19

Publication series

NameLecture Notes in Electrical Engineering
Volume330
ISSN (Print)18761100
ISSN (Electronic)18761119

Other

Other6th FTRA International Conference on Computer Science and its Applications, CSA 2014
CountryUnited States
CityGuam
Period14/12/1714/12/19

Fingerprint

Parallel processing systems
Gages
Managers
Industry

Keywords

  • Benchmark
  • Cluster management
  • HPC
  • Scheduler
  • Supercomputer

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Yoon, J. W., Hong, T. Y., Park, C. Y., & Yu, H. (2015). Stable HPC cluster management scheme through performance evaluation. In Lecture Notes in Electrical Engineering (Vol. 330, pp. 1017-1023). (Lecture Notes in Electrical Engineering; Vol. 330). Springer Verlag. https://doi.org/10.1007/978-3-662-45402-2_144

Stable HPC cluster management scheme through performance evaluation. / Yoon, Jun Weon; Hong, Tae Yeong; Park, Chan Yeol; Yu, Heonchang.

Lecture Notes in Electrical Engineering. Vol. 330 Springer Verlag, 2015. p. 1017-1023 (Lecture Notes in Electrical Engineering; Vol. 330).

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

Yoon, JW, Hong, TY, Park, CY & Yu, H 2015, Stable HPC cluster management scheme through performance evaluation. in Lecture Notes in Electrical Engineering. vol. 330, Lecture Notes in Electrical Engineering, vol. 330, Springer Verlag, pp. 1017-1023, 6th FTRA International Conference on Computer Science and its Applications, CSA 2014, Guam, United States, 14/12/17. https://doi.org/10.1007/978-3-662-45402-2_144
Yoon JW, Hong TY, Park CY, Yu H. Stable HPC cluster management scheme through performance evaluation. In Lecture Notes in Electrical Engineering. Vol. 330. Springer Verlag. 2015. p. 1017-1023. (Lecture Notes in Electrical Engineering). https://doi.org/10.1007/978-3-662-45402-2_144
Yoon, Jun Weon ; Hong, Tae Yeong ; Park, Chan Yeol ; Yu, Heonchang. / Stable HPC cluster management scheme through performance evaluation. Lecture Notes in Electrical Engineering. Vol. 330 Springer Verlag, 2015. pp. 1017-1023 (Lecture Notes in Electrical Engineering).
@inproceedings{b102a0bed98d48288057ec91245e3dd0,
title = "Stable HPC cluster management scheme through performance evaluation",
abstract = "HPC is representative tools for performing large-scale scientific calculation both academia and industry. Tachyon is a high-performance parallel computing system which constructed based on SUN Blade X6275. It composed of 3,200 computing nodes and infra-facilities. Also, this machine works with various software stacks such as file system, archive manager, compiler, debugger, parallel tools, etc. In this paper, we handle the requirements and requisites to build and manage an HPC cluster environment. In addition, we analyzed the history of batch job which include information performed by scheduler. By doing so, we are able to gauge the needs and performance of the next system to be introduced.",
keywords = "Benchmark, Cluster management, HPC, Scheduler, Supercomputer",
author = "Yoon, {Jun Weon} and Hong, {Tae Yeong} and Park, {Chan Yeol} and Heonchang Yu",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-662-45402-2_144",
language = "English",
isbn = "9783662454015",
volume = "330",
series = "Lecture Notes in Electrical Engineering",
publisher = "Springer Verlag",
pages = "1017--1023",
booktitle = "Lecture Notes in Electrical Engineering",

}

TY - GEN

T1 - Stable HPC cluster management scheme through performance evaluation

AU - Yoon, Jun Weon

AU - Hong, Tae Yeong

AU - Park, Chan Yeol

AU - Yu, Heonchang

PY - 2015/1/1

Y1 - 2015/1/1

N2 - HPC is representative tools for performing large-scale scientific calculation both academia and industry. Tachyon is a high-performance parallel computing system which constructed based on SUN Blade X6275. It composed of 3,200 computing nodes and infra-facilities. Also, this machine works with various software stacks such as file system, archive manager, compiler, debugger, parallel tools, etc. In this paper, we handle the requirements and requisites to build and manage an HPC cluster environment. In addition, we analyzed the history of batch job which include information performed by scheduler. By doing so, we are able to gauge the needs and performance of the next system to be introduced.

AB - HPC is representative tools for performing large-scale scientific calculation both academia and industry. Tachyon is a high-performance parallel computing system which constructed based on SUN Blade X6275. It composed of 3,200 computing nodes and infra-facilities. Also, this machine works with various software stacks such as file system, archive manager, compiler, debugger, parallel tools, etc. In this paper, we handle the requirements and requisites to build and manage an HPC cluster environment. In addition, we analyzed the history of batch job which include information performed by scheduler. By doing so, we are able to gauge the needs and performance of the next system to be introduced.

KW - Benchmark

KW - Cluster management

KW - HPC

KW - Scheduler

KW - Supercomputer

UR - http://www.scopus.com/inward/record.url?scp=84915749712&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84915749712&partnerID=8YFLogxK

U2 - 10.1007/978-3-662-45402-2_144

DO - 10.1007/978-3-662-45402-2_144

M3 - Conference contribution

AN - SCOPUS:84915749712

SN - 9783662454015

VL - 330

T3 - Lecture Notes in Electrical Engineering

SP - 1017

EP - 1023

BT - Lecture Notes in Electrical Engineering

PB - Springer Verlag

ER -