Title | ||
---|---|---|
Accelerating high-order mesh optimisation with an architecture-independent programming model. |
Abstract | ||
---|---|---|
Heterogeneous manycore performance-portable programming models and libraries, such as Kokkos, have been developed to facilitate portability and maintainability of high-performance computing codes and enhance their resilience to architectural changes. Here we investigate the suitability of the Kokkos programming model for optimising the performance of the high-order mesh generator NekMesh, which has been developed to efficiently generate meshes containing millions of elements for industrial problem involving complex geometries. We describe the variational approach for a posteriori high-order mesh optimisation employed within NekMesh and its parallel implementation. We discuss its implementation for modern manycore massively parallel shared-memory CPU and GPU platforms using Kokkos and demonstrate that we achieve increased performance on multicore CPUs and accelerators compared with a native Pthreads implementation. Further, we show that we achieve additional speedup and cost reduction by running on GPUs without any hardware-specific code optimisation. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1016/j.cpc.2018.03.025 | Computer Physics Communications |
Keywords | Field | DocType |
High-order mesh optimisation,Architecture-independent programming model,Kokkos,Portability,Parallel hardware,Variational framework | Mathematical optimization,Polygon mesh,Programming paradigm,Massively parallel,Parallel computing,POSIX Threads,Software portability,Multi-core processor,Mathematics,Maintainability,Speedup | Journal |
Volume | ISSN | Citations |
229 | 0010-4655 | 1 |
PageRank | References | Authors |
0.36 | 7 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jan Eichstädt | 1 | 1 | 0.36 |
Mashy Green | 2 | 1 | 0.36 |
Michael Turner | 3 | 7 | 1.43 |
Joaquim Peiró | 4 | 39 | 7.28 |
David Moxey | 5 | 9 | 3.02 |