Smoothers with localized residual computations for geometric multigrid methods
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912438928015360 |
|---|---|
| author | Wichrowski, Michał Munch, Peter Kronbichler, Martin Kanschat, Guido |
| author_facet | Wichrowski, Michał Munch, Peter Kronbichler, Martin Kanschat, Guido |
| contents | We improve the performance of multigrid solvers on many-core architectures with cache hierarchies by reorganizing operations in the smoothing step to minimize memory transfers. We focus on patch smoothers, which offer robust convergence rates with respect to the finite element degree for various equations, in the setting of multiplicative subspace correction for numerical efficiency. By combining the computation of local residuals with local solvers, we increase the locality of the problem and thus reduce data transfers. The thread-parallel implementation of this algorithm is based on coloring, which contradicts cache efficiency. We improve data locality by rearranging the loop into batches so that more data can be reused. The organization of consecutive batches prioritizes data locality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_02100 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Smoothers with localized residual computations for geometric multigrid methods Wichrowski, Michał Munch, Peter Kronbichler, Martin Kanschat, Guido Numerical Analysis We improve the performance of multigrid solvers on many-core architectures with cache hierarchies by reorganizing operations in the smoothing step to minimize memory transfers. We focus on patch smoothers, which offer robust convergence rates with respect to the finite element degree for various equations, in the setting of multiplicative subspace correction for numerical efficiency. By combining the computation of local residuals with local solvers, we increase the locality of the problem and thus reduce data transfers. The thread-parallel implementation of this algorithm is based on coloring, which contradicts cache efficiency. We improve data locality by rearranging the loop into batches so that more data can be reused. The organization of consecutive batches prioritizes data locality. |
| title | Smoothers with localized residual computations for geometric multigrid methods |
| topic | Numerical Analysis |
| url | https://arxiv.org/abs/2407.02100 |