On the performance of molecular dynamics applications on current high-end systems

Joachim Hein; F  Reid; L  Smith; I  Bush; M  Guest; P  Sherwood

doi:10.1098/rsta.2005.1624

On the performance of molecular dynamics applications on current high-end systems

Joachim Hein, F Reid, L Smith, I Bush, M Guest, P Sherwood

School of Physics and Astronomy

Research output: Contribution to journal › Article › peer-review

Abstract

The effective exploitation of current high performance computing (HPC) platforms in molecular simulation relies on the ability of the present generation of parallel molecular dynamics code to make effective ritilisation of these platforms and their components, including CPUs and memory. In this paper, we investigate the efficiency and scaling of a series of popular molecular dynamics codes on the UK's national HPC resources, an IBM p690+ cluster and an SGI Altix 3700.

Focusing primarily on the AMBER, DL_Poly and NAMD simulation codes, we demonstrate the major performance and scalabilitv advantages that arise through a distributed, rather than a replicated data approach.

Original language	English
Pages (from-to)	1987-1998
Number of pages	12
Journal	Philosophical Transactions A: Mathematical, Physical and Engineering Sciences
Volume	363
Issue number	1833
DOIs	https://doi.org/10.1098/rsta.2005.1624
Publication status	Published - 15 Aug 2005

Keywords / Materials (for Non-textual outputs)

molecular dynamics
capability computing
high performance computing
SIMULATION
PACKAGE
ENERGY

Access to Document

10.1098/rsta.2005.1624

Cite this

@article{525aaf90062d4268a9d77296bdf7785f,

title = "On the performance of molecular dynamics applications on current high-end systems",

abstract = "The effective exploitation of current high performance computing (HPC) platforms in molecular simulation relies on the ability of the present generation of parallel molecular dynamics code to make effective ritilisation of these platforms and their components, including CPUs and memory. In this paper, we investigate the efficiency and scaling of a series of popular molecular dynamics codes on the UK's national HPC resources, an IBM p690+ cluster and an SGI Altix 3700.Focusing primarily on the AMBER, DL_Poly and NAMD simulation codes, we demonstrate the major performance and scalabilitv advantages that arise through a distributed, rather than a replicated data approach.",

keywords = "molecular dynamics, capability computing, high performance computing, SIMULATION, PACKAGE, ENERGY",

author = "Joachim Hein and F Reid and L Smith and I Bush and M Guest and P Sherwood",

year = "2005",

month = aug,

day = "15",

doi = "10.1098/rsta.2005.1624",

language = "English",

volume = "363",

pages = "1987--1998",

journal = "Philosophical Transactions A: Mathematical, Physical and Engineering Sciences",

issn = "1364-503X",

publisher = "The Royal Society",

number = "1833",

}

TY - JOUR

T1 - On the performance of molecular dynamics applications on current high-end systems

AU - Hein, Joachim

AU - Reid, F

AU - Smith, L

AU - Bush, I

AU - Guest, M

AU - Sherwood, P

PY - 2005/8/15

Y1 - 2005/8/15

N2 - The effective exploitation of current high performance computing (HPC) platforms in molecular simulation relies on the ability of the present generation of parallel molecular dynamics code to make effective ritilisation of these platforms and their components, including CPUs and memory. In this paper, we investigate the efficiency and scaling of a series of popular molecular dynamics codes on the UK's national HPC resources, an IBM p690+ cluster and an SGI Altix 3700.Focusing primarily on the AMBER, DL_Poly and NAMD simulation codes, we demonstrate the major performance and scalabilitv advantages that arise through a distributed, rather than a replicated data approach.

AB - The effective exploitation of current high performance computing (HPC) platforms in molecular simulation relies on the ability of the present generation of parallel molecular dynamics code to make effective ritilisation of these platforms and their components, including CPUs and memory. In this paper, we investigate the efficiency and scaling of a series of popular molecular dynamics codes on the UK's national HPC resources, an IBM p690+ cluster and an SGI Altix 3700.Focusing primarily on the AMBER, DL_Poly and NAMD simulation codes, we demonstrate the major performance and scalabilitv advantages that arise through a distributed, rather than a replicated data approach.

KW - molecular dynamics

KW - capability computing

KW - high performance computing

KW - SIMULATION

KW - PACKAGE

KW - ENERGY

U2 - 10.1098/rsta.2005.1624

DO - 10.1098/rsta.2005.1624

M3 - Article

SN - 1364-503X

VL - 363

SP - 1987

EP - 1998

JO - Philosophical Transactions A: Mathematical, Physical and Engineering Sciences

JF - Philosophical Transactions A: Mathematical, Physical and Engineering Sciences

IS - 1833

ER -

On the performance of molecular dynamics applications on current high-end systems

Abstract

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this