Accelerating the Conjugate Gradient Algorithm with GPUs in CFD Simulations

Hartwig Anzt (1) , Marc Baboulin (2) , Jack J. Dongarra (1) , Yvan Fournier (3) , Frank Hülsemann (4) , Amal Khabou (2) , Yushan Wang (5)

1 ICL - Innovative Computing Laboratory [Knoxville]
2 ParSys - LRI - Systèmes parallèles (LRI)
3 EDF - EDF
4 EDF R&D - EDF R&D
5 LRI - Laboratoire de Recherche en Informatique

This paper illustrates how GPU computing can be used to accelerate computational fluid dynamics (CFD) simulations. For sparse linear systems arising from finite volume discretization, we evaluate and optimize the performance of Conjugate Gradient (CG) routines designed for manycore accelerators and compare against an industrial CPU-based implementation. We also investigate how the recent advances in preconditioning, such as iterative Incomplete Cholesky (IC, as symmetric case of ILU) preconditioning, match the requirements for solving real world problems.

Domaines: Analyse numérique [cs.NA]; Calcul parallèle, distribué et partagé [cs.DC]

Hartwig Anzt, Marc Baboulin, Jack J. Dongarra, Yvan Fournier, Frank Hülsemann, et al.. Accelerating the Conjugate Gradient Algorithm with GPUs in CFD Simulations. VECPAR 2016 - 12th International Meeting on High Performance Computing for Computational Science, Jun 2016, Porto, Portugal. ⟨10.1007/978-3-319-61982-8_5⟩. ⟨hal-01662043⟩