USING GRAPHICS PROCESSORS TO SOLVE SPARSE SLAE BY PRECONDITIONED KRYLOV SUBSPACE ITERATIVE METHODS AS SHOWN BY THE EX |
| 1 | |
| 2011 |
| scientific article | 517.958:532 | ||
| 205-212 | parallel computing, Krylov subspace iterative methods, graphics processors, control volume method, systems of linear algebraic equations |
| The library implementation of preconditioned Krylov subspace iterative methods on modern NVIDIA GPUs to solve large sparse systems of linear algebraic equations (SLAE) is presented. The library is intended for users who want to avoid the difficulties and details of parallel programming on GPUs, but who have to deal with large sparse SLAE and who need to use the performance of GPU parallel architecture. |
 |