A comparison of the shared-memory parallel programming models OpenMP, OpenACC and Kokkos in the context of implicit solvers for high-order FEM
| Autor: | Martin Vymazal, Jan Eichstädt, Joaquim Peiró, David Moxey |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Technology
General Physics and Astronomy Context (language use) Parallel computing Computational fluid dynamics 01 natural sciences 010305 fluids & plasmas Physics::Fluid Dynamics Shared-memory parallel programming models 0103 physical sciences Helmholtz equation 010306 general physics 01 Mathematical Sciences ComputingMethodologies_COMPUTERGRAPHICS FEM Science & Technology 02 Physical Sciences business.industry Physics OpenMP BENCHMARK PERFORMANCE Solver FRAMEWORK Data structure Nuclear & Particles Physics Finite element method Physics Mathematical Kokkos OpenACC Shared memory Hardware and Architecture Physical Sciences Computer Science Programming paradigm Benchmark (computing) Computer Science Interdisciplinary Applications 08 Information and Computing Sciences business |
| Zdroj: | Computer Physics Communications. 255:107245 |
| ISSN: | 0010-4655 |
| DOI: | 10.1016/j.cpc.2020.107245 |
| Popis: | We consider the application of three performance-portable programming models in the context of a high-order spectral element, implicit time-stepping solver for the Navier–Stokes equations. We aim to evaluate whether the use of these models allows code developers to deliver high-performance solvers for computational fluid dynamics simulations that are capable of effectively utilising both many-core CPU and GPU architectures. Using the core elliptic solver for the Navier–Stokes equations as a benchmarking guide, we evaluate the performance of these models on a range of unstructured meshes and give guidelines for the translation of existing codebases and their data structures to these models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect