Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures

Seyong Lee; John Gounley; Amanda Randles; Jeffrey S. Vetter

doi:10.1016/j.jpdc.2019.02.005

Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures

Seyong Lee, John Gounley, Amanda Randles, Jeffrey S. Vetter

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

15 Scopus citations

Abstract

Patient-specific hemodynamic simulations have the potential to greatly improve both the diagnosis and treatment of a variety of vascular diseases. Portability will enable wider adoption of computational fluid dynamics (CFD) applications in the biomedical research community and targeting to platforms ideally suited to different vascular regions. In this work, we present a case study in performance portability that assesses (1) the ease of porting an MPI application optimized for one specific architecture to new platforms using variants of hybrid MPI＋X programming models; (2) performance portability seen when simulating blood flow in three different vascular regions on diverse heterogeneous architectures; (3) model-based performance prediction for future architectures; and (4) performance scaling of the hybrid MPI＋X programming on parallel heterogeneous systems. We discuss the lessons learned in porting HARVEY, a massively parallel CFD application, from traditional multicore CPUs to diverse heterogeneous architectures ranging from NVIDIA/AMD GPUs to Intel MICs and Altera FPGAs.

Original language	English
Pages (from-to)	1-13
Number of pages	13
Journal	Journal of Parallel and Distributed Computing
Volume	129
DOIs	https://doi.org/10.1016/j.jpdc.2019.02.005
State	Published - Jul 2019

Funding

This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory (ORNL), which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725, and computing facility that came from the Lawrence Livermore National Laboratory (LLNL) Institutional Computing Grand Challenge program. This material is based upon work supported by the U.S. Department of Energy (DOE), Office of Science, Office of Advanced Scientific Computing Research, United States. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the DOE. The United States Government (USG) retains and the publisher, by accepting the article for publication, acknowledges that the USG retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for USG purposes. The DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). Research reported in this publication is also supported by the ORNL Joint Faculty Program, United States and the Office of the Director, National Institutes of Health, United States under Award Number DP5OD019876. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory (ORNL), which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725 , and computing facility that came from the Lawrence Livermore National Laboratory (LLNL) Institutional Computing Grand Challenge program. This material is based upon work supported by the U.S. Department of Energy (DOE), Office of Science, Office of Advanced Scientific Computing Research, United States . This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the DOE. The United States Government (USG) retains and the publisher, by accepting the article for publication, acknowledges that the USG retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for USG purposes. The DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan ( http://energy.gov/downloads/doe-public-access-plan ). Research reported in this publication is also supported by the ORNL Joint Faculty Program, United States and the Office of the Director, National Institutes of Health, United States under Award Number DP5OD019876 . The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Funders	Funder number
National Institutes of Health	DP5OD019876
U.S. Department of Energy	DE-AC05-00OR22725
Office of the Director
Office of Science
Advanced Scientific Computing Research
Oak Ridge National Laboratory

Keywords

Computational fluid dynamics
Heterogeneous architectures
Lattice Boltzmann method
OpenACC
Patient-specific hemodynamics
Performance portability
Performance prediction

Access to Document

10.1016/j.jpdc.2019.02.005

Cite this

@article{4b68b2790c704a5cabb4ddbe26baa7a6,

title = "Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures",

abstract = "Patient-specific hemodynamic simulations have the potential to greatly improve both the diagnosis and treatment of a variety of vascular diseases. Portability will enable wider adoption of computational fluid dynamics (CFD) applications in the biomedical research community and targeting to platforms ideally suited to different vascular regions. In this work, we present a case study in performance portability that assesses (1) the ease of porting an MPI application optimized for one specific architecture to new platforms using variants of hybrid MPI＋X programming models; (2) performance portability seen when simulating blood flow in three different vascular regions on diverse heterogeneous architectures; (3) model-based performance prediction for future architectures; and (4) performance scaling of the hybrid MPI＋X programming on parallel heterogeneous systems. We discuss the lessons learned in porting HARVEY, a massively parallel CFD application, from traditional multicore CPUs to diverse heterogeneous architectures ranging from NVIDIA/AMD GPUs to Intel MICs and Altera FPGAs.",

keywords = "Computational fluid dynamics, Heterogeneous architectures, Lattice Boltzmann method, OpenACC, Patient-specific hemodynamics, Performance portability, Performance prediction",

author = "Seyong Lee and John Gounley and Amanda Randles and Vetter, {Jeffrey S.}",

note = "Publisher Copyright: {\textcopyright} 2019 Elsevier Inc.",

year = "2019",

month = jul,

doi = "10.1016/j.jpdc.2019.02.005",

language = "English",

volume = "129",

pages = "1--13",

journal = "Journal of Parallel and Distributed Computing",

issn = "0743-7315",

publisher = "Elsevier",

}

TY - JOUR

T1 - Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures

AU - Lee, Seyong

AU - Gounley, John

AU - Randles, Amanda

AU - Vetter, Jeffrey S.

PY - 2019/7

Y1 - 2019/7

N2 - Patient-specific hemodynamic simulations have the potential to greatly improve both the diagnosis and treatment of a variety of vascular diseases. Portability will enable wider adoption of computational fluid dynamics (CFD) applications in the biomedical research community and targeting to platforms ideally suited to different vascular regions. In this work, we present a case study in performance portability that assesses (1) the ease of porting an MPI application optimized for one specific architecture to new platforms using variants of hybrid MPI＋X programming models; (2) performance portability seen when simulating blood flow in three different vascular regions on diverse heterogeneous architectures; (3) model-based performance prediction for future architectures; and (4) performance scaling of the hybrid MPI＋X programming on parallel heterogeneous systems. We discuss the lessons learned in porting HARVEY, a massively parallel CFD application, from traditional multicore CPUs to diverse heterogeneous architectures ranging from NVIDIA/AMD GPUs to Intel MICs and Altera FPGAs.

AB - Patient-specific hemodynamic simulations have the potential to greatly improve both the diagnosis and treatment of a variety of vascular diseases. Portability will enable wider adoption of computational fluid dynamics (CFD) applications in the biomedical research community and targeting to platforms ideally suited to different vascular regions. In this work, we present a case study in performance portability that assesses (1) the ease of porting an MPI application optimized for one specific architecture to new platforms using variants of hybrid MPI＋X programming models; (2) performance portability seen when simulating blood flow in three different vascular regions on diverse heterogeneous architectures; (3) model-based performance prediction for future architectures; and (4) performance scaling of the hybrid MPI＋X programming on parallel heterogeneous systems. We discuss the lessons learned in porting HARVEY, a massively parallel CFD application, from traditional multicore CPUs to diverse heterogeneous architectures ranging from NVIDIA/AMD GPUs to Intel MICs and Altera FPGAs.

KW - Computational fluid dynamics

KW - Heterogeneous architectures

KW - Lattice Boltzmann method

KW - OpenACC

KW - Patient-specific hemodynamics

KW - Performance portability

KW - Performance prediction

UR - http://www.scopus.com/inward/record.url?scp=85063080601&partnerID=8YFLogxK

U2 - 10.1016/j.jpdc.2019.02.005

DO - 10.1016/j.jpdc.2019.02.005

M3 - Article

AN - SCOPUS:85063080601

SN - 0743-7315

VL - 129

SP - 1

EP - 13

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

ER -

Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this