Two-Stage Gauss-Seidel Preconditioners and Smoothers for Krylov Solvers on a GPU Cluster: Preprint
Conference
·
OSTI ID:1845268
Gauss-Seidel (GS) relaxation is often employed as a preconditioner for a Krylov solver or as a smoother for Algebraic Multigrid (AMG). However, the requisite sparse triangular solve is difficult to parallelize on many-core architectures such as graphics processing units (GPUs). In the present study, the performance of the sequential GS relaxation based on a triangular solve is compared with two-stage variants, replacing the direct triangular solve with a fixed number of inner Jacobi-Richardson (JR) iterations. When a small number of inner iterations is sufficient to maintain the Krylov convergence rate, the two-stage GS (GS2) often outperforms the sequential algorithm on many-core architectures. The GS2 algorithm is also compared with JR. When they perform the same number of ops for SpMV (e.g. three JR sweeps compared to two GS sweeps with one inner JR sweep), the GS2 iterations, and the Krylov solver preconditioned with GS2, may converge faster than the JR iterations. Moreover, for some problems (e.g. elasticity), it was found that JR may diverge with a damping factor of one, whereas two-stage GS may improve the convergence with more inner iterations. Finally, to study the performance of the two-stage smoother and preconditioner for a practical problem, these were applied to incompressible uid ow simulations on GPUs.
- Research Organization:
- National Renewable Energy Laboratory (NREL), Golden, CO (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
- DOE Contract Number:
- AC36-08GO28308;
- OSTI ID:
- 1845268
- Report Number(s):
- NREL/CP-2C00-80263; MainId:42466; UUID:1736ac04-4292-4d20-b918-e0bfb6997b92; MainAdminID:63010
- Conference Information:
- Presented at the 2021 SIAM Annual Meeting, 19-23 July 2021
- Country of Publication:
- United States
- Language:
- English
Similar Records
Neumann Series in MGS-GMRES and Inner-Outer Iterations: Preprint
Scaled ILU Smoothers for Navier-Stokes Pressure Projection
Multigrid Smoothers for Ultra-Parallel Computing: Additional Theory and Discussion
Conference
·
Tue Feb 08 19:00:00 EST 2022
·
OSTI ID:1845270
Scaled ILU Smoothers for Navier-Stokes Pressure Projection
Journal Article
·
Wed Dec 27 23:00:00 EST 2023
· International Journal for Numerical Methods in Fluids
·
OSTI ID:2377647
Multigrid Smoothers for Ultra-Parallel Computing: Additional Theory and Discussion
Technical Report
·
Tue Jun 14 00:00:00 EDT 2011
·
OSTI ID:1122232