A proof of concept for very fast finite element Poisson solvers on accelerator hardware
| Autor: | Dustin Ruda, Stefan Turek, Dirk Ribbrock, Peter Zajac |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: | |
| DOI: | 10.17877/de290r-22630 |
| Popis: | It is demonstrated that modern accelerator hardware specialized in AI, e.g., ���next gen GPUs��� equipped with Tensor Cores, can be profitably used in finite element simulations by means of a new hardware-oriented method to solve linear systems arising from Poisson's equation in 2D. We consider the NVIDIA Tesla V100 Tensor Core GPU with a peak performance of 125 TFLOP/s, that is only achievable in half precision and if operations with high arithmetic intensity, such as dense matrix multiplications, are executed, though. Its computing power can be exploited to a great extent by the new method based on ���prehandling��� without loss of accuracy. We obtain a significant reduction of computing time compared to a standard geometric multigrid solver on standard x64 hardware. Proceedings in applied mathematics and mechanics;21(1) |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|