Application-aware management of parallel simulation collections

Siu Man Yau, Vijay Karamcheti, Denis Zorin, Kostadin Damevski, Steven G. Parker

Research output: Contribution to journalArticle

Abstract

This paper presents a system deployed on parallel clusters to manage a collection of parallel simulations that make up a computational study. It explores how such a system can extend traditional parallel job scheduling and resource allocation techniques to incorporate knowledge specific to the study. Using a UINTAH-based helium gas simulation code (ARCHES) and the SimX system for multi-experiment computational studies, this paper demonstrates that, by using application-specific knowledge in resource allocation and scheduling decisions, one can reduce the run time of a computational study from over 20 hours to under 4.5 hours on a 32-processor cluster, and from almost 11 hours to just over 3.5 hours on a 64-processor cluster.

Original languageEnglish (US)
Pages (from-to)35-44
Number of pages10
JournalACM SIGPLAN Notices
Volume44
Issue number4
StatePublished - 2009

Fingerprint

Resource allocation
Scheduling
Helium
Gases
Experiments

Keywords

  • High-throughput computing
  • Parallel system

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Yau, S. M., Karamcheti, V., Zorin, D., Damevski, K., & Parker, S. G. (2009). Application-aware management of parallel simulation collections. ACM SIGPLAN Notices, 44(4), 35-44.

Application-aware management of parallel simulation collections. / Yau, Siu Man; Karamcheti, Vijay; Zorin, Denis; Damevski, Kostadin; Parker, Steven G.

In: ACM SIGPLAN Notices, Vol. 44, No. 4, 2009, p. 35-44.

Research output: Contribution to journalArticle

Yau, SM, Karamcheti, V, Zorin, D, Damevski, K & Parker, SG 2009, 'Application-aware management of parallel simulation collections', ACM SIGPLAN Notices, vol. 44, no. 4, pp. 35-44.
Yau SM, Karamcheti V, Zorin D, Damevski K, Parker SG. Application-aware management of parallel simulation collections. ACM SIGPLAN Notices. 2009;44(4):35-44.
Yau, Siu Man ; Karamcheti, Vijay ; Zorin, Denis ; Damevski, Kostadin ; Parker, Steven G. / Application-aware management of parallel simulation collections. In: ACM SIGPLAN Notices. 2009 ; Vol. 44, No. 4. pp. 35-44.
@article{963d79a2cd174edb92c89a4004e500d4,
title = "Application-aware management of parallel simulation collections",
abstract = "This paper presents a system deployed on parallel clusters to manage a collection of parallel simulations that make up a computational study. It explores how such a system can extend traditional parallel job scheduling and resource allocation techniques to incorporate knowledge specific to the study. Using a UINTAH-based helium gas simulation code (ARCHES) and the SimX system for multi-experiment computational studies, this paper demonstrates that, by using application-specific knowledge in resource allocation and scheduling decisions, one can reduce the run time of a computational study from over 20 hours to under 4.5 hours on a 32-processor cluster, and from almost 11 hours to just over 3.5 hours on a 64-processor cluster.",
keywords = "High-throughput computing, Parallel system",
author = "Yau, {Siu Man} and Vijay Karamcheti and Denis Zorin and Kostadin Damevski and Parker, {Steven G.}",
year = "2009",
language = "English (US)",
volume = "44",
pages = "35--44",
journal = "ACM SIGPLAN Notices",
issn = "1523-2867",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

TY - JOUR

T1 - Application-aware management of parallel simulation collections

AU - Yau, Siu Man

AU - Karamcheti, Vijay

AU - Zorin, Denis

AU - Damevski, Kostadin

AU - Parker, Steven G.

PY - 2009

Y1 - 2009

N2 - This paper presents a system deployed on parallel clusters to manage a collection of parallel simulations that make up a computational study. It explores how such a system can extend traditional parallel job scheduling and resource allocation techniques to incorporate knowledge specific to the study. Using a UINTAH-based helium gas simulation code (ARCHES) and the SimX system for multi-experiment computational studies, this paper demonstrates that, by using application-specific knowledge in resource allocation and scheduling decisions, one can reduce the run time of a computational study from over 20 hours to under 4.5 hours on a 32-processor cluster, and from almost 11 hours to just over 3.5 hours on a 64-processor cluster.

AB - This paper presents a system deployed on parallel clusters to manage a collection of parallel simulations that make up a computational study. It explores how such a system can extend traditional parallel job scheduling and resource allocation techniques to incorporate knowledge specific to the study. Using a UINTAH-based helium gas simulation code (ARCHES) and the SimX system for multi-experiment computational studies, this paper demonstrates that, by using application-specific knowledge in resource allocation and scheduling decisions, one can reduce the run time of a computational study from over 20 hours to under 4.5 hours on a 32-processor cluster, and from almost 11 hours to just over 3.5 hours on a 64-processor cluster.

KW - High-throughput computing

KW - Parallel system

UR - http://www.scopus.com/inward/record.url?scp=70350614751&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350614751&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:70350614751

VL - 44

SP - 35

EP - 44

JO - ACM SIGPLAN Notices

JF - ACM SIGPLAN Notices

SN - 1523-2867

IS - 4

ER -