Implementation of a nonsymmetric linear solver on GPU
Numerical methods and programming, Tome 10 (2009) no. 3, pp. 321-326.

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

An implementation of a nonsymmetric sparse linear solver on GPU is considered. The solver uses a version of the BiCG-STAB algorithm. This algorithm is shortly described. Several sparse matrix storage formats are given with consideration of the NVIDIA GPGPU hardware features. The performance of the implementation is analyzed compared to the performance of a similar algorithm run on contemporary CPUs. The impact of several factors on the performance is discussed with some suggestions on further development.
Keywords: parallel computing; sparse linear solver; GPGPU; CUDA; BiCG-STAB.
@article{VMP_2009_10_3_a6,
     author = {S. N. Chadov},
     title = {Implementation of a nonsymmetric linear solver on {GPU}},
     journal = {Numerical methods and programming},
     pages = {321--326},
     publisher = {mathdoc},
     volume = {10},
     number = {3},
     year = {2009},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2009_10_3_a6/}
}
TY  - JOUR
AU  - S. N. Chadov
TI  - Implementation of a nonsymmetric linear solver on GPU
JO  - Numerical methods and programming
PY  - 2009
SP  - 321
EP  - 326
VL  - 10
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2009_10_3_a6/
LA  - ru
ID  - VMP_2009_10_3_a6
ER  - 
%0 Journal Article
%A S. N. Chadov
%T Implementation of a nonsymmetric linear solver on GPU
%J Numerical methods and programming
%D 2009
%P 321-326
%V 10
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2009_10_3_a6/
%G ru
%F VMP_2009_10_3_a6
S. N. Chadov. Implementation of a nonsymmetric linear solver on GPU. Numerical methods and programming, Tome 10 (2009) no. 3, pp. 321-326. http://geodesic.mathdoc.fr/item/VMP_2009_10_3_a6/