NekRS, a GPU-accelerated spectral element Navier–Stokes solver

Paul Fischer; Stefan Kerkemeier; Misun Min; Yu Hsiang Lan; Malachi Phillips; Thilina Rathnayake; Elia Merzari; Ananias Tomboulides; Ali Karakus; Noel Chalmers; Tim Warburton

doi:10.1016/j.parco.2022.102982

NekRS, a GPU-accelerated spectral element Navier–Stokes solver

Paul Fischer, Stefan Kerkemeier, Misun Min, Yu Hsiang Lan, Malachi Phillips, Thilina Rathnayake, Elia Merzari, Ananias Tomboulides, Ali Karakus, Noel Chalmers, Tim Warburton

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

Abstract

The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B grid points (240B degrees-of-freedom).

Original language	English (US)
Article number	102982
Journal	Parallel Computing
Volume	114
DOIs	https://doi.org/10.1016/j.parco.2022.102982
State	Published - Dec 2022

Keywords

Exascale applications
GPU
Incompressible Navier–Stokes
LibParanumal
Nek5000
NekRS
OCCA
Performance
Scalability
Spectral element method

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Computer Graphics and Computer-Aided Design
Artificial Intelligence

Online availability

10.1016/j.parco.2022.102982

Library availability

Discover UIUC Full Text

Cite this

@article{90439c0adaa141a28360e35c6a5741f2,

title = "NekRS, a GPU-accelerated spectral element Navier–Stokes solver",

abstract = "The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B grid points (240B degrees-of-freedom).",

keywords = "Exascale applications, GPU, Incompressible Navier–Stokes, LibParanumal, Nek5000, NekRS, OCCA, Performance, Scalability, Spectral element method",

author = "Paul Fischer and Stefan Kerkemeier and Misun Min and Lan, {Yu Hsiang} and Malachi Phillips and Thilina Rathnayake and Elia Merzari and Ananias Tomboulides and Ali Karakus and Noel Chalmers and Tim Warburton",

note = "Publisher Copyright: {\textcopyright} 2022 Argonne National Laboratory, The Author(s)",

year = "2022",

month = dec,

doi = "10.1016/j.parco.2022.102982",

language = "English (US)",

volume = "114",

journal = "Parallel Computing",

issn = "0167-8191",

publisher = "Elsevier",

}

TY - JOUR

T1 - NekRS, a GPU-accelerated spectral element Navier–Stokes solver

AU - Fischer, Paul

AU - Kerkemeier, Stefan

AU - Min, Misun

AU - Lan, Yu Hsiang

AU - Phillips, Malachi

AU - Rathnayake, Thilina

AU - Merzari, Elia

AU - Tomboulides, Ananias

AU - Karakus, Ali

AU - Chalmers, Noel

AU - Warburton, Tim

PY - 2022/12

Y1 - 2022/12

N2 - The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B grid points (240B degrees-of-freedom).

AB - The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B grid points (240B degrees-of-freedom).

KW - Exascale applications

KW - GPU

KW - Incompressible Navier–Stokes

KW - LibParanumal

KW - Nek5000

KW - NekRS

KW - OCCA

KW - Performance

KW - Scalability

KW - Spectral element method

UR - http://www.scopus.com/inward/record.url?scp=85141764266&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85141764266&partnerID=8YFLogxK

U2 - 10.1016/j.parco.2022.102982

DO - 10.1016/j.parco.2022.102982

M3 - Article

AN - SCOPUS:85141764266

SN - 0167-8191

VL - 114

JO - Parallel Computing

JF - Parallel Computing

M1 - 102982

ER -

NekRS, a GPU-accelerated spectral element Navier–Stokes solver

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this