Pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics simulations

Samuel Schlachter, Stephen Herbein, Shuching Ou, Jeremy S. Logan, Sandeep Patel, Michela Taufer

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

To address the challenge of pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics (MD) simulations on nondedicated high-end clusters, our work aims to supplement, rather than rewrite, existing workflow and resource managers. To this end, we propose a companion module that complements workflow managers and a wrapper module that supports resource managers. We model the maximum utilization of our approach in comparison to the traditional common approach for two molecular simulations: a sodium dodecyl sulfate (SDS) system with dynamically variable job runtimes and a carbon nanotube system with hardware and application failures. In light of our solution, we estimate increased utilization in both simulations. These findings implicitly assure a more coordinated progression of long trajectories across GPUs on a cluster's nodes.

Original languageEnglish (US)
Article number6616581
Pages (from-to)40-50
Number of pages11
JournalIEEE Design and Test
Volume31
Issue number1
DOIs
StatePublished - Feb 2014
Externally publishedYes

Fingerprint

Molecular dynamics
Managers
Trajectories
Computer simulation
Sodium dodecyl sulfate
Carbon nanotubes
Hardware
Graphics processing unit

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Electrical and Electronic Engineering

Cite this

Pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics simulations. / Schlachter, Samuel; Herbein, Stephen; Ou, Shuching; Logan, Jeremy S.; Patel, Sandeep; Taufer, Michela.

In: IEEE Design and Test, Vol. 31, No. 1, 6616581, 02.2014, p. 40-50.

Research output: Contribution to journalArticle

Schlachter, Samuel ; Herbein, Stephen ; Ou, Shuching ; Logan, Jeremy S. ; Patel, Sandeep ; Taufer, Michela. / Pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics simulations. In: IEEE Design and Test. 2014 ; Vol. 31, No. 1. pp. 40-50.
@article{b4230546577e448ea53f1b88f10191d4,
title = "Pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics simulations",
abstract = "To address the challenge of pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics (MD) simulations on nondedicated high-end clusters, our work aims to supplement, rather than rewrite, existing workflow and resource managers. To this end, we propose a companion module that complements workflow managers and a wrapper module that supports resource managers. We model the maximum utilization of our approach in comparison to the traditional common approach for two molecular simulations: a sodium dodecyl sulfate (SDS) system with dynamically variable job runtimes and a carbon nanotube system with hardware and application failures. In light of our solution, we estimate increased utilization in both simulations. These findings implicitly assure a more coordinated progression of long trajectories across GPUs on a cluster's nodes.",
author = "Samuel Schlachter and Stephen Herbein and Shuching Ou and Logan, {Jeremy S.} and Sandeep Patel and Michela Taufer",
year = "2014",
month = "2",
doi = "10.1109/MDAT.2013.2284203",
language = "English (US)",
volume = "31",
pages = "40--50",
journal = "IEEE Design and Test",
issn = "2168-2356",
publisher = "IEEE Computer Society",
number = "1",

}

TY - JOUR

T1 - Pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics simulations

AU - Schlachter, Samuel

AU - Herbein, Stephen

AU - Ou, Shuching

AU - Logan, Jeremy S.

AU - Patel, Sandeep

AU - Taufer, Michela

PY - 2014/2

Y1 - 2014/2

N2 - To address the challenge of pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics (MD) simulations on nondedicated high-end clusters, our work aims to supplement, rather than rewrite, existing workflow and resource managers. To this end, we propose a companion module that complements workflow managers and a wrapper module that supports resource managers. We model the maximum utilization of our approach in comparison to the traditional common approach for two molecular simulations: a sodium dodecyl sulfate (SDS) system with dynamically variable job runtimes and a carbon nanotube system with hardware and application failures. In light of our solution, we estimate increased utilization in both simulations. These findings implicitly assure a more coordinated progression of long trajectories across GPUs on a cluster's nodes.

AB - To address the challenge of pursuing coordinated trajectory progression and efficient resource utilization of GPU-enabled molecular dynamics (MD) simulations on nondedicated high-end clusters, our work aims to supplement, rather than rewrite, existing workflow and resource managers. To this end, we propose a companion module that complements workflow managers and a wrapper module that supports resource managers. We model the maximum utilization of our approach in comparison to the traditional common approach for two molecular simulations: a sodium dodecyl sulfate (SDS) system with dynamically variable job runtimes and a carbon nanotube system with hardware and application failures. In light of our solution, we estimate increased utilization in both simulations. These findings implicitly assure a more coordinated progression of long trajectories across GPUs on a cluster's nodes.

UR - http://www.scopus.com/inward/record.url?scp=84896758813&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896758813&partnerID=8YFLogxK

U2 - 10.1109/MDAT.2013.2284203

DO - 10.1109/MDAT.2013.2284203

M3 - Article

AN - SCOPUS:84896758813

VL - 31

SP - 40

EP - 50

JO - IEEE Design and Test

JF - IEEE Design and Test

SN - 2168-2356

IS - 1

M1 - 6616581

ER -