Abstract
The main contribution of the present work consists of several parallel approaches for grid refinement based on a multi-domain decomposition for lattice-Boltzmann simulations. The proposed method for discretizing the fluid incorporates different regular Cartesian grids with no homogeneous spatial domains, which are in need to be communicated each other. Three different parallel approaches are proposed, homogeneous Multicore, homogeneous GPU, and heterogeneous Multicore-GPU. Although, the homogeneous implementations exhibit satisfactory results, the heterogeneous approach achieves up to 30% extra efficiency, in terms of Millions of Fluid Lattice Updates per Second (MFLUPS), by overlapping some of the steps on both architectures, Multicore and GPU.
Original language | English |
---|---|
Title of host publication | Proceedings - IEEE 18th International Conference on Computational Science and Engineering, CSE 2015 |
Editors | Christian Plessl, Guojing Cong, Joao M. P. Cardoso, Luis Veiga, Didier El Baz, Thomas Rauber |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 1-8 |
Number of pages | 8 |
ISBN (Electronic) | 9781467382977 |
DOIs | |
State | Published - 2015 |
Externally published | Yes |
Event | 18th IEEE International Conference on Computational Science and Engineering, CSE 2015 - Porto, Portugal Duration: Oct 21 2015 → Oct 23 2015 |
Publication series
Name | Proceedings - IEEE 18th International Conference on Computational Science and Engineering, CSE 2015 |
---|
Conference
Conference | 18th IEEE International Conference on Computational Science and Engineering, CSE 2015 |
---|---|
Country/Territory | Portugal |
City | Porto |
Period | 10/21/15 → 10/23/15 |
Funding
This research has been supported by EU-FET grant EUNISON 308874, the Basque Excellence Research Center (BERC 2014-2017) program by the Basque Government, the Spanish Ministry of Economy and Competitiveness MINECO: BCAM Severo Ochoa accreditation SEV-2013-0323 and the Project of the Spanish Ministry of Economy and Competitiveness with reference MTM2013-40824. We also thank NVIDIA GPU Research Center program for the provided resources.
Keywords
- Heterogeneous (Multicore-GPU) Platforms
- Lattice-Boltzmann Method
- Multi-Domain Grid Refinement
- Parallel Computing