PeleC: An adaptive mesh refinement solver for compressible reacting flows

被引:54
作者
de Frahan, Marc T. Henry [1 ]
Rood, Jon S. [1 ]
Day, Marc S. [1 ]
Sitaraman, Hariswaran [1 ]
Yellapantula, Shashank [1 ]
Perry, Bruce A. [1 ]
Grout, Ray W. [1 ]
Almgren, Ann [2 ]
Zhang, Weiqun [2 ]
Bell, John B. [2 ]
Chen, Jacqueline H. [3 ]
机构
[1] Natl Renewable Energy Lab, Computat Sci Ctr, 15013 Denver W Pkwy, Golden, CO 80401 USA
[2] Lawrence Berkeley Natl Lab, Ctr Computat Sci & Engn, Berkeley, CA USA
[3] Sandia Natl Labs, Combust Res Facil, Livermore, CA USA
关键词
High performance computing; graphics processing units; combustion; computational fluid dynamics; compressible reacting flows; adaptive mesh refinement; NUMERICAL-SIMULATION; HYDRODYNAMICS; VERIFICATION; SCHEMES; CODE;
D O I
10.1177/10943420221121151
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Reacting flow simulations for combustion applications require extensive computing capabilities. Leveraging the AMReX library, the Pele suite of combustion simulation tools targets the largest supercomputers available and future exascale machines. We introduce PeleC, the compressible solver in the Pele suite, and detail its capabilities, including complex geometry representation, chemistry integration, and discretization. We present a comparison of development efforts using both OpenACC and AMReX's C++ performance portability framework for execution on multiple GPU architectures. We discuss relevant details that have allowed PeleC to achieve high performance and scalability. PeleC's performance characteristics are measured through relevant simulations on multiple supercomputers. The success of PeleC's design for exascale is exhibited through demonstration of a 160 billion cell simulation and weak scaling onto 100% of Summit, an NVIDIA-based GPU supercomputer at Oak Ridge National Laboratory. Our results provide confidence that PeleC will enable future combustion science simulations with unprecedented fidelity.
引用
收藏
页码:115 / 131
页数:17
相关论文
共 57 条
[1]  
Advanced Micro Devices, 2022, HIP PROGR GUID
[2]   CASTRO: A NEW COMPRESSIBLE ASTROPHYSICAL SOLVER. I. HYDRODYNAMICS AND SELF-GRAVITY [J].
Almgren, A. S. ;
Beckner, V. E. ;
Bell, B. ;
Day, M. S. ;
Howell, L. H. ;
Joggerst, C. C. ;
Lijewski, M. J. ;
Nonaka, A. ;
Singer, M. ;
Zingale, M. .
ASTROPHYSICAL JOURNAL, 2010, 715 (02) :1221-1238
[3]  
Almgren A. S., 2022, AMREX HYDRO
[4]   A conservative adaptive projection method for the variable density incompressible Navier-Stokes equations [J].
Almgren, AS ;
Bell, JB ;
Colella, P ;
Howell, LH ;
Welcome, ML .
JOURNAL OF COMPUTATIONAL PHYSICS, 1998, 142 (01) :1-46
[5]  
AMReX Team, 2022, AMREXS DOC
[6]  
[Anonymous], 2013, TILLING DURABLE ABST
[7]  
Argonne Leadership Computing Facility, 2022, THET
[8]   RAJA: Portable Performance for Large-Scale Scientific Applications [J].
Beckingsale, David Alexander ;
Burmark, Jason ;
Hornung, Rich ;
Jones, Holger ;
Killian, William ;
Kunen, Adam J. ;
Pearce, Olga ;
Robinson, Peter ;
Ryujin, Brian S. ;
Scogland, Thomas R. W. .
PROCEEDINGS OF P3HPC 2019: 2019 IEEE/ACM INTERNATIONAL WORKSHOP ON PERFORMANCE, PORTABILITY AND PRODUCTIVITY IN HPC (P3HPC), 2019, :71-81
[9]   Numerical simulation of a laboratory-scale turbulent slot flame [J].
Bell, John B. ;
Day, Marcus S. ;
Grcar, Joseph F. ;
Lijewski, Michael J. ;
Driscoll, James F. ;
Filatyev, Sergel A. .
PROCEEDINGS OF THE COMBUSTION INSTITUTE, 2007, 31 :1299-1307
[10]   A state redistribution algorithm for finite volume schemes on cut cell meshes [J].
Berger, Marsha ;
Giuliani, Andrew .
JOURNAL OF COMPUTATIONAL PHYSICS, 2021, 428