Parallel Computational Modelling of Inelastic Neutron Scattering in Multi-node and Multi-core Architectures

H. Gonzales-Velez; M. T. Garba; D. L. Roach

doi:10.1109/HPCC.2010.45

Parallel Computational Modelling of Inelastic Neutron Scattering in Multi-node and Multi-core Architectures

H. Gonzales-Velez, M. T. Garba, D. L. Roach

School of Informatics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper examines the initial parallel implementation of SCATTER, a computationally intensive inelastic neutron scattering routine with polycrystalline averaging capability, for the General Utility Lattice Program (GULP). Of particular importance to structural investigation on the atomic scale, this work identifies the computational features of SCATTER relevant to a parallel implementation and presents initial results from performance tests on multi-core and multi-node environments. Our initial approach exhibits near-linear scalability up to 256 MPI processes for a significant model.

Original language	English
Title of host publication	The 11th IEEE International Conference on High Performance Computing and Communications (HPCC-09)
Pages	509-514
Number of pages	6
DOIs	https://doi.org/10.1109/HPCC.2010.45
Publication status	Published - 26 Nov 2010

Keywords / Materials (for Non-textual outputs)

Algorithmic Skeletons
Computational Physics
Lattice Dynamics
Multi-Core Processors
Neutron Scattering
Parallel Programming
Scientific computing

Access to Document

10.1109/HPCC.2010.45

Cite this

@inproceedings{833b04e5e6074b81aee00459a99a7236,

title = "Parallel Computational Modelling of Inelastic Neutron Scattering in Multi-node and Multi-core Architectures",

abstract = "This paper examines the initial parallel implementation of SCATTER, a computationally intensive inelastic neutron scattering routine with polycrystalline averaging capability, for the General Utility Lattice Program (GULP). Of particular importance to structural investigation on the atomic scale, this work identifies the computational features of SCATTER relevant to a parallel implementation and presents initial results from performance tests on multi-core and multi-node environments. Our initial approach exhibits near-linear scalability up to 256 MPI processes for a significant model.",

keywords = "Algorithmic Skeletons, Computational Physics, Lattice Dynamics, Multi-Core Processors, Neutron Scattering, Parallel Programming, Scientific computing",

author = "H. Gonzales-Velez and Garba, \{M. T.\} and Roach, \{D. L.\}",

year = "2010",

month = nov,

day = "26",

doi = "10.1109/HPCC.2010.45",

language = "English",

isbn = "978-1-4244-8335-8",

pages = "509--514",

booktitle = "The 11th IEEE International Conference on High Performance Computing and Communications (HPCC-09)",

}

TY - GEN

T1 - Parallel Computational Modelling of Inelastic Neutron Scattering in Multi-node and Multi-core Architectures

AU - Gonzales-Velez, H.

AU - Garba, M. T.

AU - Roach, D. L.

PY - 2010/11/26

Y1 - 2010/11/26

N2 - This paper examines the initial parallel implementation of SCATTER, a computationally intensive inelastic neutron scattering routine with polycrystalline averaging capability, for the General Utility Lattice Program (GULP). Of particular importance to structural investigation on the atomic scale, this work identifies the computational features of SCATTER relevant to a parallel implementation and presents initial results from performance tests on multi-core and multi-node environments. Our initial approach exhibits near-linear scalability up to 256 MPI processes for a significant model.

AB - This paper examines the initial parallel implementation of SCATTER, a computationally intensive inelastic neutron scattering routine with polycrystalline averaging capability, for the General Utility Lattice Program (GULP). Of particular importance to structural investigation on the atomic scale, this work identifies the computational features of SCATTER relevant to a parallel implementation and presents initial results from performance tests on multi-core and multi-node environments. Our initial approach exhibits near-linear scalability up to 256 MPI processes for a significant model.

KW - Algorithmic Skeletons

KW - Computational Physics

KW - Lattice Dynamics

KW - Multi-Core Processors

KW - Neutron Scattering

KW - Parallel Programming

KW - Scientific computing

U2 - 10.1109/HPCC.2010.45

DO - 10.1109/HPCC.2010.45

M3 - Conference contribution

SN - 978-1-4244-8335-8

SP - 509

EP - 514

BT - The 11th IEEE International Conference on High Performance Computing and Communications (HPCC-09)

ER -

Parallel Computational Modelling of Inelastic Neutron Scattering in Multi-node and Multi-core Architectures

Abstract

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this