2013
DOI: 10.1016/j.compfluid.2012.01.025
|View full text |Cite
|
Sign up to set email alerts
|

Towards a complete FEM-based simulation toolkit on GPUs: Unstructured grid finite element geometric multigrid solvers with strong smoothers based on sparse approximate inverses

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
19
0

Year Published

2014
2014
2020
2020

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 26 publications
(19 citation statements)
references
References 16 publications
0
19
0
Order By: Relevance
“…In continuation of the first project phase, we enhanced the assembly of sparse approximate inverses (SPAI), a kind of preconditioner that we had shown to be very effective within the DUNE solver before [26,9]. Concerning the assembly of such matrices we have investigated three strategies regarding their numerical efficacy (that is their quality in approximating A −1 ), the computational complexity of the actual assembly and ultimately, the total efficiency of the amortised assembly combined with all applications during a system solution.…”
Section: Strong Smoothers On the Gpu: Fast Approximate Inverses With mentioning
confidence: 99%
“…In continuation of the first project phase, we enhanced the assembly of sparse approximate inverses (SPAI), a kind of preconditioner that we had shown to be very effective within the DUNE solver before [26,9]. Concerning the assembly of such matrices we have investigated three strategies regarding their numerical efficacy (that is their quality in approximating A −1 ), the computational complexity of the actual assembly and ultimately, the total efficiency of the amortised assembly combined with all applications during a system solution.…”
Section: Strong Smoothers On the Gpu: Fast Approximate Inverses With mentioning
confidence: 99%
“…A GPU approach for geometric multigrid solvers on finite elements for unstructured grid problems was performed by Geveler et al . , in which their GPU implementation is based on cascades of sparse matrix‐vector multiplication by applying strong smoothers. Komatitsch et al have used CUDA to speedup numerical simulation of seismic wave propagation resulting from earthquakes.…”
Section: Related Work On Graphics Processor Units and High Performancmentioning
confidence: 99%
“…Additionally, the Jacobi method depends on the sparse matrix-vector multiplication operation, which has low computational density and is generally memory bandwidth bounded. In [31, 32, 33], the authors introduce GPU-based linear solvers with multigrid methods. The solvers use the ELLpack sparse matrix data structure for their specific problems, which is not efficient when number of non-zero entries per row varies largely.…”
Section: Previous Workmentioning
confidence: 99%