|
Numerical methods and programming, 2009, Volume 10, Issue 3, Pages 321–326
(Mi vmp384)
|
|
|
|
Вычислительные методы и приложения
Implementation of a nonsymmetric linear solver on GPU
S. N. Chadov Ivanovo State Power University
Abstract:
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.
Citation:
S. N. Chadov, “Implementation of a nonsymmetric linear solver on GPU”, Num. Meth. Prog., 10:3 (2009), 321–326
Linking options:
https://www.mathnet.ru/eng/vmp384 https://www.mathnet.ru/eng/vmp/v10/i3/p321
|
Statistics & downloads: |
Abstract page: | 83 | Full-text PDF : | 36 |
|