Towards extreme-scale simulations for low mach fluids with second-generation trilinos

Paul Lin; Matthew Bettencourt; Stefan Domino; Travis Fisher; Mark Hoemmen; Jonathan Hu; Eric Phipps; Andrey Prokopenko; Sivasankaran Rajamanickam; Christopher Siefert; Stephen Kennon

doi:10.1142/S0129626414420055

Towards extreme-scale simulations for low mach fluids with second-generation trilinos

Paul Lin, Matthew Bettencourt, Stefan Domino, Travis Fisher, Mark Hoemmen, Jonathan Hu, Eric Phipps, Andrey Prokopenko, Sivasankaran Rajamanickam, Christopher Siefert, Stephen Kennon

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

29 Scopus citations

Abstract

Trilinos is an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. While Trilinos was originally designed for scalable solutions of large problems, the fidelity needed by many simulations is significantly greater than what one could have envisioned two decades ago. When problem sizes exceed a billion elements even scalable applications and solver stacks require a complete revision. The second-generation Trilinos employs C++ templates in order to solve arbitrarily large problems. We present a case study of the integration of Trilinos with a low Mach fluids engineering application (SIERRA low Mach module/Nalu). Through the use of improved algorithms and better software engineering practices, we demonstrate good weak scaling for up to a nine billion element large eddy simulation (LES) problem on unstructured meshes with a 27 billion row matrix on 524,288 cores of an IBM Blue Gene/Q platform.

Original language	English
Article number	1442005
Journal	Parallel processing letters
Volume	24
Issue number	4
DOIs	https://doi.org/10.1142/S0129626414420055
State	Published - Dec 22 2014
Externally published	Yes

Funding

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lock-heed Martin Company, for the United States Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

Funders	Funder number
Sandia Corporation
National Nuclear Security Administration	DE-AC04-94AL85000
Sandia National Laboratories

Keywords

Solver library
Trilinos
extreme-scale simulations
vertical integration

Access to Document

10.1142/S0129626414420055

Cite this

@article{5152e78fd548497b8af3d6edbd38a1de,

title = "Towards extreme-scale simulations for low mach fluids with second-generation trilinos",

abstract = "Trilinos is an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. While Trilinos was originally designed for scalable solutions of large problems, the fidelity needed by many simulations is significantly greater than what one could have envisioned two decades ago. When problem sizes exceed a billion elements even scalable applications and solver stacks require a complete revision. The second-generation Trilinos employs C++ templates in order to solve arbitrarily large problems. We present a case study of the integration of Trilinos with a low Mach fluids engineering application (SIERRA low Mach module/Nalu). Through the use of improved algorithms and better software engineering practices, we demonstrate good weak scaling for up to a nine billion element large eddy simulation (LES) problem on unstructured meshes with a 27 billion row matrix on 524,288 cores of an IBM Blue Gene/Q platform.",

keywords = "Solver library, Trilinos, extreme-scale simulations, vertical integration",

author = "Paul Lin and Matthew Bettencourt and Stefan Domino and Travis Fisher and Mark Hoemmen and Jonathan Hu and Eric Phipps and Andrey Prokopenko and Sivasankaran Rajamanickam and Christopher Siefert and Stephen Kennon",

note = "Publisher Copyright: {\textcopyright} 2014 World Scientific Publishing Company.",

year = "2014",

month = dec,

day = "22",

doi = "10.1142/S0129626414420055",

language = "English",

volume = "24",

journal = "Parallel processing letters",

issn = "0129-6264",

publisher = "World Scientific Publishing",

number = "4",

}

TY - JOUR

T1 - Towards extreme-scale simulations for low mach fluids with second-generation trilinos

AU - Lin, Paul

AU - Bettencourt, Matthew

AU - Domino, Stefan

AU - Fisher, Travis

AU - Hoemmen, Mark

AU - Hu, Jonathan

AU - Phipps, Eric

AU - Prokopenko, Andrey

AU - Rajamanickam, Sivasankaran

AU - Siefert, Christopher

AU - Kennon, Stephen

PY - 2014/12/22

Y1 - 2014/12/22

N2 - Trilinos is an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. While Trilinos was originally designed for scalable solutions of large problems, the fidelity needed by many simulations is significantly greater than what one could have envisioned two decades ago. When problem sizes exceed a billion elements even scalable applications and solver stacks require a complete revision. The second-generation Trilinos employs C++ templates in order to solve arbitrarily large problems. We present a case study of the integration of Trilinos with a low Mach fluids engineering application (SIERRA low Mach module/Nalu). Through the use of improved algorithms and better software engineering practices, we demonstrate good weak scaling for up to a nine billion element large eddy simulation (LES) problem on unstructured meshes with a 27 billion row matrix on 524,288 cores of an IBM Blue Gene/Q platform.

AB - Trilinos is an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. While Trilinos was originally designed for scalable solutions of large problems, the fidelity needed by many simulations is significantly greater than what one could have envisioned two decades ago. When problem sizes exceed a billion elements even scalable applications and solver stacks require a complete revision. The second-generation Trilinos employs C++ templates in order to solve arbitrarily large problems. We present a case study of the integration of Trilinos with a low Mach fluids engineering application (SIERRA low Mach module/Nalu). Through the use of improved algorithms and better software engineering practices, we demonstrate good weak scaling for up to a nine billion element large eddy simulation (LES) problem on unstructured meshes with a 27 billion row matrix on 524,288 cores of an IBM Blue Gene/Q platform.

KW - Solver library

KW - Trilinos

KW - extreme-scale simulations

KW - vertical integration

UR - http://www.scopus.com/inward/record.url?scp=84929199220&partnerID=8YFLogxK

U2 - 10.1142/S0129626414420055

DO - 10.1142/S0129626414420055

M3 - Article

AN - SCOPUS:84929199220

SN - 0129-6264

VL - 24

JO - Parallel processing letters

JF - Parallel processing letters

IS - 4

M1 - 1442005

ER -

Towards extreme-scale simulations for low mach fluids with second-generation trilinos

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this