Parallel Triangular Solvers on GPU
University of Calgary
University of Calgary, 2012
@article{liu2012parallel,
title={Parallel Triangular Solvers on GPU},
author={Liu, H. and Yu, S. and Chen, Z.},
year={2012}
}
In this paper, we investigate GPU based parallel triangular solvers systematically. The parallel triangular solvers are fundamental to incomplete LU factorization family preconditioners and algebraic multigrid solvers. We develop a new matrix format suitable for GPU devices. Parallel lower triangular solver and upper triangular solver are developed for this new data structure. With these solvers, ILU preconditioner and domain decomposition preconditioner are developed. Numerical results show that Krylov subspace solver with these parallel triangular solvers are around 10 times faster than our CPU based solvers.
June 3, 2012 by hgpu
