Loading...
Loading...

Go to the content (press return)

Portable implementation model for CFD simulations. Application to hybrid CPU/GPU supercomputers

Author
Oyarzun, G.; Borrell, R.; Gorobets, A.; Oliva, A.
Type of activity
Journal article
Journal
International journal of computational fluid dynamics
Date of publication
2017-10-27
Volume
31
Number
9
First page
396
Last page
411
DOI
https://doi.org/10.1080/10618562.2017.1390084 Open in new window
Repository
http://hdl.handle.net/2117/121221 Open in new window
URL
http://www.tandfonline.com/doi/full/10.1080/10618562.2017.1390084 Open in new window
Abstract
Nowadays, high performance computing (HPC) systems experience a disruptive moment with a variety of novel architectures and frameworks, without any clarity of which one is going to prevail. In this context, the portability of codes across different architectures is of major importance. This paper presents a portable implementation model based on an algebraic operational approach for direct numerical simulation (DNS) and large eddy simulation (LES) of incompressible turbulent flows using unstruct...
Citation
Oyarzun, G., Borrell, R., Gorobets, A., Oliva, A. Portable implementation model for CFD simulations. Application to hybrid CPU/GPU supercomputers. "International journal of computational fluid dynamics", 27 Octubre 2017, vol. 31, núm. 9, p. 396-411.
Keywords
CFD code, GPU, Heterogeneous computing, MPI+CUDA, OpenCL, Portability, Sliced ELLPACK
Group of research
CTTC - Heat and Mass Transfer Technological Center

Participants

Attachments