Iterative Sparse Triangular Solves for Preconditioning [chapter]

Hartwig Anzt, Edmond Chow, Jack Dongarra
2015 Lecture Notes in Computer Science  
Sparse triangular solvers are typically parallelized using levelscheduling techniques, but parallel efficiency is poor on high-throughput architectures like GPUs. We propose using an iterative approach for solving sparse triangular systems when an approximation is suitable. This approach will not work for all problems, but can be successful for sparse triangular matrices arising from incomplete factorizations, where an approximate solution is acceptable. We demonstrate the performance gains
more » ... this approach can have on GPUs in the context of solving sparse linear systems with a preconditioned Krylov subspace method. We also illustrate the effect of using asynchronous iterations.
doi:10.1007/978-3-662-48096-0_50 fatcat:k4p5ysqnhza3djruia4kvsc5j4