On peculiarities of using heterogeneous cluster architecture for
Numerical methods and programming, Tome 12 (2011) no. 4, pp. 450-460
Cet article a éte moissonné depuis la source Math-Net.Ru
The use of high-performance computing systems built on the basis of NVIDIA graphics processors is discussed. Several models of parallelism are compared, for using in the iterative Krylov subspace methods intended for solving large sparse systems of linear algebraic equations that arise in the numerical solution of a wide class of continuum mechanics problems described by partial differential equations. The numerical computations are based on the application of such technologies as MPI, OpenMP, Boost, and CUDA. The proposed methods are tested on the supercomputer "GraphIT!" installed at the Research Computing Center of Moscow State University when solving a problem of nonstationary flow of a fluid to the wells with complex trajectories in a three-dimensional domain.
Keywords:
parallel computing; graphics processors; iterative Krylov subspace methods; method of control volumes; flow in porous media.
@article{VMP_2011_12_4_a7,
author = {D. A. Gubaidullin and A. I. Nikiforov and R. Sadovnikov},
title = {On peculiarities of using heterogeneous cluster architecture for},
journal = {Numerical methods and programming},
pages = {450--460},
year = {2011},
volume = {12},
number = {4},
language = {ru},
url = {http://geodesic.mathdoc.fr/item/VMP_2011_12_4_a7/}
}
TY - JOUR AU - D. A. Gubaidullin AU - A. I. Nikiforov AU - R. Sadovnikov TI - On peculiarities of using heterogeneous cluster architecture for JO - Numerical methods and programming PY - 2011 SP - 450 EP - 460 VL - 12 IS - 4 UR - http://geodesic.mathdoc.fr/item/VMP_2011_12_4_a7/ LA - ru ID - VMP_2011_12_4_a7 ER -
D. A. Gubaidullin; A. I. Nikiforov; R. Sadovnikov. On peculiarities of using heterogeneous cluster architecture for. Numerical methods and programming, Tome 12 (2011) no. 4, pp. 450-460. http://geodesic.mathdoc.fr/item/VMP_2011_12_4_a7/