Parallelization of the FLAPW method

A. Canning, W. Mannstadt, Arthur J Freeman

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

The FLAPW (full-potential linearized-augmented plane-wave) method is one of the most accurate first-principles methods for determining structural, electronic and magnetic properties of crystals and surfaces. Until the present work, the FLAPW method has been limited to systems of less than about a hundred atoms due to the lack of an efficient parallel implementation to exploit the power and memory of parallel computers. In this work, we present an efficient parallelization of the method by division among the processors of the plane-wave components for each state. The code is also optimized for RISC (reduced instruction set computer) architectures, such as those found on most parallel computers, making full use of BLAS (basic linear algebra subprograms) wherever possible. Scaling results are presented for systems of up to 686 silicon atoms and 343 palladium atoms per unit cell, running on up to 512 processors on a CRAY T3E parallel supercomputer.

Original languageEnglish
Pages (from-to)233-243
Number of pages11
JournalComputer Physics Communications
Volume130
Issue number3
Publication statusPublished - Aug 15 2000

Fingerprint

parallel computers
plane waves
Atoms
central processing units
architecture (computers)
atoms
supercomputers
Linear algebra
Computer architecture
Supercomputers
Electronic properties
Palladium
division
Structural properties
palladium
Magnetic properties
algebra
magnetic properties
scaling
Data storage equipment

Keywords

  • Density functional theory
  • FLAPW
  • Full-potential LAPW
  • Parallelization
  • Total energy

ASJC Scopus subject areas

  • Computer Science Applications
  • Physics and Astronomy(all)

Cite this

Canning, A., Mannstadt, W., & Freeman, A. J. (2000). Parallelization of the FLAPW method. Computer Physics Communications, 130(3), 233-243.

Parallelization of the FLAPW method. / Canning, A.; Mannstadt, W.; Freeman, Arthur J.

In: Computer Physics Communications, Vol. 130, No. 3, 15.08.2000, p. 233-243.

Research output: Contribution to journalArticle

Canning, A, Mannstadt, W & Freeman, AJ 2000, 'Parallelization of the FLAPW method', Computer Physics Communications, vol. 130, no. 3, pp. 233-243.
Canning A, Mannstadt W, Freeman AJ. Parallelization of the FLAPW method. Computer Physics Communications. 2000 Aug 15;130(3):233-243.
Canning, A. ; Mannstadt, W. ; Freeman, Arthur J. / Parallelization of the FLAPW method. In: Computer Physics Communications. 2000 ; Vol. 130, No. 3. pp. 233-243.
@article{90ffa9925eae47579e27a7c57036250b,
title = "Parallelization of the FLAPW method",
abstract = "The FLAPW (full-potential linearized-augmented plane-wave) method is one of the most accurate first-principles methods for determining structural, electronic and magnetic properties of crystals and surfaces. Until the present work, the FLAPW method has been limited to systems of less than about a hundred atoms due to the lack of an efficient parallel implementation to exploit the power and memory of parallel computers. In this work, we present an efficient parallelization of the method by division among the processors of the plane-wave components for each state. The code is also optimized for RISC (reduced instruction set computer) architectures, such as those found on most parallel computers, making full use of BLAS (basic linear algebra subprograms) wherever possible. Scaling results are presented for systems of up to 686 silicon atoms and 343 palladium atoms per unit cell, running on up to 512 processors on a CRAY T3E parallel supercomputer.",
keywords = "Density functional theory, FLAPW, Full-potential LAPW, Parallelization, Total energy",
author = "A. Canning and W. Mannstadt and Freeman, {Arthur J}",
year = "2000",
month = "8",
day = "15",
language = "English",
volume = "130",
pages = "233--243",
journal = "Computer Physics Communications",
issn = "0010-4655",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Parallelization of the FLAPW method

AU - Canning, A.

AU - Mannstadt, W.

AU - Freeman, Arthur J

PY - 2000/8/15

Y1 - 2000/8/15

N2 - The FLAPW (full-potential linearized-augmented plane-wave) method is one of the most accurate first-principles methods for determining structural, electronic and magnetic properties of crystals and surfaces. Until the present work, the FLAPW method has been limited to systems of less than about a hundred atoms due to the lack of an efficient parallel implementation to exploit the power and memory of parallel computers. In this work, we present an efficient parallelization of the method by division among the processors of the plane-wave components for each state. The code is also optimized for RISC (reduced instruction set computer) architectures, such as those found on most parallel computers, making full use of BLAS (basic linear algebra subprograms) wherever possible. Scaling results are presented for systems of up to 686 silicon atoms and 343 palladium atoms per unit cell, running on up to 512 processors on a CRAY T3E parallel supercomputer.

AB - The FLAPW (full-potential linearized-augmented plane-wave) method is one of the most accurate first-principles methods for determining structural, electronic and magnetic properties of crystals and surfaces. Until the present work, the FLAPW method has been limited to systems of less than about a hundred atoms due to the lack of an efficient parallel implementation to exploit the power and memory of parallel computers. In this work, we present an efficient parallelization of the method by division among the processors of the plane-wave components for each state. The code is also optimized for RISC (reduced instruction set computer) architectures, such as those found on most parallel computers, making full use of BLAS (basic linear algebra subprograms) wherever possible. Scaling results are presented for systems of up to 686 silicon atoms and 343 palladium atoms per unit cell, running on up to 512 processors on a CRAY T3E parallel supercomputer.

KW - Density functional theory

KW - FLAPW

KW - Full-potential LAPW

KW - Parallelization

KW - Total energy

UR - http://www.scopus.com/inward/record.url?scp=0000622986&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000622986&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0000622986

VL - 130

SP - 233

EP - 243

JO - Computer Physics Communications

JF - Computer Physics Communications

SN - 0010-4655

IS - 3

ER -