Stencil-aware GPU optimization of iterative solvers
From MaRDI portal
\texttt{PETSc}autotuningfinite-differenceGPGPUiterative solversNewton-Krylov methodnumerical examplessparse matrix formatstructured grid
Nonlinear elliptic equations (35J60) Iterative numerical methods for linear systems (65F10) Computational methods for sparse matrices (65F50) Finite difference methods for boundary value problems involving PDEs (65N06) Mesh generation, refinement, and adaptive methods for boundary value problems involving PDEs (65N50)
Recommendations
- Algebraic Multigrid Using a Stencil–CSR Hybrid Format on GPUs
- Accelerating iterative linear solvers using multiple graphical processing units
- Accelerating preconditioned iterative linear solvers on GPU
- scientific article; zbMATH DE number 1759987
- Pipelined iterative solvers with kernel fusion for graphics processing units
Cited in
(7)- Code modernization strategies to 3-D stencil-based applications on intel Xeon Phi: KNC and KNL
- A new memory mapping mechanism for GPGPUs' stencil computation
- Architecture and performance of Devito, a system for automated stencil computation
- Accelerating stencil computation on GPGPU by novel mapping method between the global memory and the shared memory
- Stencil scaling for vector-valued PDEs on hybrid grids with applications to generalized Newtonian fluids
- Pipelined iterative solvers with kernel fusion for graphics processing units
- Low Byte/Flop Implementation of Iterative Solver for Sparse Matrices Derived from Stencil Computations
This page was built for publication: Stencil-aware GPU optimization of iterative solvers
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2870688)
