Generating Coupling Interfaces for Multiphysics Simulations with ExaStencils and waLBerla

被引:1
|
作者
Angersbach, Richard [1 ]
Kuckuk, Sebastian [2 ]
Koestler, Harald [2 ]
机构
[1] Friedrich Alexander Univ Erlangen Nurnberg, Dept Comp Sci, Erlangen, Germany
[2] Friedrich Alexander Univ Erlangen Nurnberg, Natl High Performance Comp Ctr, Nurnberg, Germany
关键词
Terms-Software Coupling; Code Generation; Multigrid; Domain-specific Language; ExaStencils; waLBerla; Lattice Boltz-mann;
D O I
10.1109/IPDPSW59300.2023.00112
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Domain-specific languages (DSLs) and code generation technology are adopted more and more by the high performance computing (HPC) community to tackle the performance, portability, and productivity requirements for HPC software. They share their inherent specialization in specific scientific domains with traditional HPC frameworks. In this regard, multiphysics simulations pose a strong challenge as they usually require components from multiple domains. A promising option for retaining these requirements is the combination of code generation with HPC frameworks through automatically created couplings. This work demonstrates the fusion of waLBerla with generated code from ExaStencils. waLBerla is a massively parallel C++ framework focused on computational fluid dynamics (CFD) simulations using the Lattice Boltzmann method (LBM). ExaStencils generates whole simulation codes using geometric multigrid solvers from specifications in its own DSL. We showcase the coupling of generated solvers with LBM simulations, including the mapping of data structures and augmentations to convert quantities units on-the-fly.
引用
收藏
页码:651 / 661
页数:11
相关论文
共 50 条
  • [1] WALBERLA: A block-structured high-performance framework for multiphysics simulations
    Bauer, Martin
    Eibl, Sebastian
    Godenschwager, Christian
    Kohl, Nils
    Kuron, Michael
    Rettinger, Christoph
    Schornbaum, Florian
    Schwarzmeier, Christoph
    Thoennes, Dominik
    Koestler, Harald
    Ruede, Ulrich
    COMPUTERS & MATHEMATICS WITH APPLICATIONS, 2021, 81 : 478 - 501
  • [2] Multilingual interfaces for parallel coupling in multiphysics and multiscale systems
    Ong, Everest T.
    Larson, J. Walter
    Norris, Boyana
    Jacob, Robert L.
    Tobis, Michael
    Steder, Michael
    COMPUTATIONAL SCIENCE - ICCS 2007, PT 1, PROCEEDINGS, 2007, 4487 : 931 - +
  • [3] An efficient parallel coupling method for multiphysics numerical simulations
    HPCC and Laboratory of Computational Physics, Institute of Applied Physics and Computational Mathematics, Beijing 100088, China
    Jisuanji Xuebao, 2007, 9 (1559-1566):
  • [4] A Python']Python extension for the massively parallel multiphysics simulation framework WALBERLA
    Bauer, Martin
    Schornbaum, Florian
    Godenschwager, Christian
    Markl, Matthias
    Anderl, Daniela
    Koestler, Harald
    Ruede, Ulrich
    INTERNATIONAL JOURNAL OF PARALLEL EMERGENT AND DISTRIBUTED SYSTEMS, 2016, 31 (06) : 529 - 542
  • [5] Towards generating efficient flow solvers with the ExaStencils approach
    Kuckuk, Sebastian
    Haase, Gundolf
    Vasco, Diego A.
    Koestler, Harald
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2017, 29 (17):
  • [6] WaLBerla: HPC software design for computational engineering simulations
    Feichtinger, C.
    Donath, S.
    Koestler, H.
    Goetz, J.
    Ruede, U.
    JOURNAL OF COMPUTATIONAL SCIENCE, 2011, 2 (02) : 105 - 112
  • [7] Volumetric coupling approaches for multiphysics simulations on non-matching meshes
    Farah, P.
    Vuong, A. -T.
    Wall, W. A.
    Popp, A.
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2016, 108 (12) : 1550 - 1576
  • [8] A discrete differential geometric formulation of multiphase surface interfaces for scalable multiphysics equilibrium simulations
    Endres, Stefan Christian
    Avila, Marc
    Maedler, Lutz
    CHEMICAL ENGINEERING SCIENCE, 2022, 257
  • [9] Multiphysics coupling of codes
    Peszynska, M
    Lu, Q
    Wheeler, MF
    COMPUTATIONAL METHODS IN WATER RESOURCES, VOLS 1 AND 2: COMPUTATIONAL METHODS FOR SUBSURFACE FLOW AND TRANSPORT - COMPUTATIONAL METHODS, SURFACE WATER SYSTEMS AND HYDROLOGY, 2000, : 175 - 182
  • [10] A coupling environment for partitioned multiphysics simulations applied to fluid-structure interaction scenarios
    Gatzhammer, Bernhard
    Mehl, Miriam
    Neckel, Tobias
    ICCS 2010 - INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE, PROCEEDINGS, 2010, 1 (01): : 681 - 689