Optimization of Finite-Differencing Kernels for Numerical Relativity Applications

Roberto Alfieri, Sebastiano Bernuzzi, Albino Perego, David Radice

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: In-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.

Original languageEnglish (US)
Title of host publicationParallel Computing is Everywhere
EditorsGerhard R. Joubert, Patrizio Dazzi, Frans Peters, Marco Danelutto, Sanzio Bassini
PublisherIOS Press BV
Pages743-749
Number of pages7
ISBN (Electronic)9781614998426
DOIs
StatePublished - Jan 1 2018

Publication series

NameAdvances in Parallel Computing
Volume32
ISSN (Print)0927-5452
ISSN (Electronic)1879-808X

Fingerprint

Relativity
Data storage equipment

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Alfieri, R., Bernuzzi, S., Perego, A., & Radice, D. (2018). Optimization of Finite-Differencing Kernels for Numerical Relativity Applications. In G. R. Joubert, P. Dazzi, F. Peters, M. Danelutto, & S. Bassini (Eds.), Parallel Computing is Everywhere (pp. 743-749). (Advances in Parallel Computing; Vol. 32). IOS Press BV. https://doi.org/10.3233/978-1-61499-843-3-743
Alfieri, Roberto ; Bernuzzi, Sebastiano ; Perego, Albino ; Radice, David. / Optimization of Finite-Differencing Kernels for Numerical Relativity Applications. Parallel Computing is Everywhere. editor / Gerhard R. Joubert ; Patrizio Dazzi ; Frans Peters ; Marco Danelutto ; Sanzio Bassini. IOS Press BV, 2018. pp. 743-749 (Advances in Parallel Computing).
@inbook{2613d77274d3455590ad750be7c05345,
title = "Optimization of Finite-Differencing Kernels for Numerical Relativity Applications",
abstract = "A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: In-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.",
author = "Roberto Alfieri and Sebastiano Bernuzzi and Albino Perego and David Radice",
year = "2018",
month = "1",
day = "1",
doi = "10.3233/978-1-61499-843-3-743",
language = "English (US)",
series = "Advances in Parallel Computing",
publisher = "IOS Press BV",
pages = "743--749",
editor = "Joubert, {Gerhard R.} and Patrizio Dazzi and Frans Peters and Marco Danelutto and Sanzio Bassini",
booktitle = "Parallel Computing is Everywhere",

}

Alfieri, R, Bernuzzi, S, Perego, A & Radice, D 2018, Optimization of Finite-Differencing Kernels for Numerical Relativity Applications. in GR Joubert, P Dazzi, F Peters, M Danelutto & S Bassini (eds), Parallel Computing is Everywhere. Advances in Parallel Computing, vol. 32, IOS Press BV, pp. 743-749. https://doi.org/10.3233/978-1-61499-843-3-743

Optimization of Finite-Differencing Kernels for Numerical Relativity Applications. / Alfieri, Roberto; Bernuzzi, Sebastiano; Perego, Albino; Radice, David.

Parallel Computing is Everywhere. ed. / Gerhard R. Joubert; Patrizio Dazzi; Frans Peters; Marco Danelutto; Sanzio Bassini. IOS Press BV, 2018. p. 743-749 (Advances in Parallel Computing; Vol. 32).

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Optimization of Finite-Differencing Kernels for Numerical Relativity Applications

AU - Alfieri, Roberto

AU - Bernuzzi, Sebastiano

AU - Perego, Albino

AU - Radice, David

PY - 2018/1/1

Y1 - 2018/1/1

N2 - A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: In-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.

AB - A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: In-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.

UR - http://www.scopus.com/inward/record.url?scp=85043605295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043605295&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-843-3-743

DO - 10.3233/978-1-61499-843-3-743

M3 - Chapter

AN - SCOPUS:85043605295

T3 - Advances in Parallel Computing

SP - 743

EP - 749

BT - Parallel Computing is Everywhere

A2 - Joubert, Gerhard R.

A2 - Dazzi, Patrizio

A2 - Peters, Frans

A2 - Danelutto, Marco

A2 - Bassini, Sanzio

PB - IOS Press BV

ER -

Alfieri R, Bernuzzi S, Perego A, Radice D. Optimization of Finite-Differencing Kernels for Numerical Relativity Applications. In Joubert GR, Dazzi P, Peters F, Danelutto M, Bassini S, editors, Parallel Computing is Everywhere. IOS Press BV. 2018. p. 743-749. (Advances in Parallel Computing). https://doi.org/10.3233/978-1-61499-843-3-743