Matches in SemOpenAlex for { <https://semopenalex.org/work/W4229063364> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4229063364 endingPage "474" @default.
- W4229063364 startingPage "459" @default.
- W4229063364 abstract "Recently, accelerator hardware in the form of graphics cards including Tensor Cores, specialized for AI, has significantly gained importance in the domain of high-performance computing. For example, NVIDIA’s Tesla V100 promises a computing power of up to 125 TFLOP/s achieved by Tensor Cores, but only if half precision floating point format is used. We describe the difficulties and discrepancy between theoretical and actual computing power if one seeks to use such hardware for numerical simulations, that is, solving partial differential equations with a matrix-based finite element method, with numerical examples. If certain requirements, namely low condition numbers and many dense matrix operations, are met, the indicated high performance can be reached without an excessive loss of accuracy. A new method to solve linear systems arising from Poisson’s equation in 2D that meets these requirements, based on “prehandling” by means of hier-archical finite elements and an additional Schur complement approach, is presented and analyzed. We provide numerical results illustrating the computational performance of this method and compare it to a commonly used (geometric) multigrid solver on standard hardware. It turns out that we can exploit nearly the full computational power of Tensor Cores and achieve a significant speed-up compared to the standard methodology without losing accuracy." @default.
- W4229063364 created "2022-05-08" @default.
- W4229063364 creator A5005126720 @default.
- W4229063364 creator A5018038936 @default.
- W4229063364 creator A5025481957 @default.
- W4229063364 creator A5059837563 @default.
- W4229063364 date "2022-05-06" @default.
- W4229063364 modified "2023-10-05" @default.
- W4229063364 title "Very fast finite element Poisson solvers on lower precision accelerator hardware: A proof of concept study for Nvidia Tesla V100" @default.
- W4229063364 cites W1506342804 @default.
- W4229063364 cites W2022397154 @default.
- W4229063364 cites W2147601343 @default.
- W4229063364 cites W2943869808 @default.
- W4229063364 cites W3108141724 @default.
- W4229063364 cites W3138530731 @default.
- W4229063364 cites W3147515309 @default.
- W4229063364 cites W3149341318 @default.
- W4229063364 cites W3158549294 @default.
- W4229063364 cites W3158805782 @default.
- W4229063364 doi "https://doi.org/10.1177/10943420221084657" @default.
- W4229063364 hasPublicationYear "2022" @default.
- W4229063364 type Work @default.
- W4229063364 citedByCount "2" @default.
- W4229063364 countsByYear W42290633642023 @default.
- W4229063364 crossrefType "journal-article" @default.
- W4229063364 hasAuthorship W4229063364A5005126720 @default.
- W4229063364 hasAuthorship W4229063364A5018038936 @default.
- W4229063364 hasAuthorship W4229063364A5025481957 @default.
- W4229063364 hasAuthorship W4229063364A5059837563 @default.
- W4229063364 hasBestOaLocation W42290633641 @default.
- W4229063364 hasConcept C11413529 @default.
- W4229063364 hasConcept C121332964 @default.
- W4229063364 hasConcept C133095886 @default.
- W4229063364 hasConcept C134306372 @default.
- W4229063364 hasConcept C135628077 @default.
- W4229063364 hasConcept C137119250 @default.
- W4229063364 hasConcept C155281189 @default.
- W4229063364 hasConcept C173608175 @default.
- W4229063364 hasConcept C199360897 @default.
- W4229063364 hasConcept C202444582 @default.
- W4229063364 hasConcept C2778770139 @default.
- W4229063364 hasConcept C28826006 @default.
- W4229063364 hasConcept C33923547 @default.
- W4229063364 hasConcept C35912277 @default.
- W4229063364 hasConcept C36503486 @default.
- W4229063364 hasConcept C41008148 @default.
- W4229063364 hasConcept C459310 @default.
- W4229063364 hasConcept C51255310 @default.
- W4229063364 hasConcept C84211073 @default.
- W4229063364 hasConcept C93779851 @default.
- W4229063364 hasConcept C97355855 @default.
- W4229063364 hasConceptScore W4229063364C11413529 @default.
- W4229063364 hasConceptScore W4229063364C121332964 @default.
- W4229063364 hasConceptScore W4229063364C133095886 @default.
- W4229063364 hasConceptScore W4229063364C134306372 @default.
- W4229063364 hasConceptScore W4229063364C135628077 @default.
- W4229063364 hasConceptScore W4229063364C137119250 @default.
- W4229063364 hasConceptScore W4229063364C155281189 @default.
- W4229063364 hasConceptScore W4229063364C173608175 @default.
- W4229063364 hasConceptScore W4229063364C199360897 @default.
- W4229063364 hasConceptScore W4229063364C202444582 @default.
- W4229063364 hasConceptScore W4229063364C2778770139 @default.
- W4229063364 hasConceptScore W4229063364C28826006 @default.
- W4229063364 hasConceptScore W4229063364C33923547 @default.
- W4229063364 hasConceptScore W4229063364C35912277 @default.
- W4229063364 hasConceptScore W4229063364C36503486 @default.
- W4229063364 hasConceptScore W4229063364C41008148 @default.
- W4229063364 hasConceptScore W4229063364C459310 @default.
- W4229063364 hasConceptScore W4229063364C51255310 @default.
- W4229063364 hasConceptScore W4229063364C84211073 @default.
- W4229063364 hasConceptScore W4229063364C93779851 @default.
- W4229063364 hasConceptScore W4229063364C97355855 @default.
- W4229063364 hasIssue "4" @default.
- W4229063364 hasLocation W42290633641 @default.
- W4229063364 hasOpenAccess W4229063364 @default.
- W4229063364 hasPrimaryLocation W42290633641 @default.
- W4229063364 hasRelatedWork W2026855204 @default.
- W4229063364 hasRelatedWork W2076361081 @default.
- W4229063364 hasRelatedWork W2112041918 @default.
- W4229063364 hasRelatedWork W2130336959 @default.
- W4229063364 hasRelatedWork W2336476964 @default.
- W4229063364 hasRelatedWork W3101543398 @default.
- W4229063364 hasRelatedWork W3217108803 @default.
- W4229063364 hasRelatedWork W4308235368 @default.
- W4229063364 hasRelatedWork W4322208717 @default.
- W4229063364 hasRelatedWork W4323043748 @default.
- W4229063364 hasVolume "36" @default.
- W4229063364 isParatext "false" @default.
- W4229063364 isRetracted "false" @default.
- W4229063364 workType "article" @default.