An efficient implementation of the lattice Boltzmann method for hybrid supercomputers
Numerical methods and programming, Tome 16 (2015) no. 2, pp. 205-214.

Voir la notice de l'article provenant de la source Math-Net.Ru

A number of features of an efficient implementation of the lattice Boltzmann method (LBM) for hybrid supercomputers with many graphics processing units (GPU) are discussed. The main strategies for reducing the memory space required by LBM are described. The performance dependence of the implemented solver on the number of the GPUs in use is analyzed for the Lomonosov supercomputer installed at Moscow State University.
Keywords: Lattice Boltzmann method, LBM, cuda, multi-gpu, high-performance computing, graphics processing unit, lattice Boltzmann method, CUDA, multi-gpu, scalability.
@article{VMP_2015_16_2_a3,
     author = {D. A. Bikulov},
     title = {An efficient implementation of the lattice {Boltzmann} method for hybrid supercomputers},
     journal = {Numerical methods and programming},
     pages = {205--214},
     publisher = {mathdoc},
     volume = {16},
     number = {2},
     year = {2015},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2015_16_2_a3/}
}
TY  - JOUR
AU  - D. A. Bikulov
TI  - An efficient implementation of the lattice Boltzmann method for hybrid supercomputers
JO  - Numerical methods and programming
PY  - 2015
SP  - 205
EP  - 214
VL  - 16
IS  - 2
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2015_16_2_a3/
LA  - ru
ID  - VMP_2015_16_2_a3
ER  - 
%0 Journal Article
%A D. A. Bikulov
%T An efficient implementation of the lattice Boltzmann method for hybrid supercomputers
%J Numerical methods and programming
%D 2015
%P 205-214
%V 16
%N 2
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2015_16_2_a3/
%G ru
%F VMP_2015_16_2_a3
D. A. Bikulov. An efficient implementation of the lattice Boltzmann method for hybrid supercomputers. Numerical methods and programming, Tome 16 (2015) no. 2, pp. 205-214. http://geodesic.mathdoc.fr/item/VMP_2015_16_2_a3/