cuIBM — A GPU-accelerated Immersed Boundary Method
Department of Mechanical Engineering, Boston University, Boston, MA, 02215, USA
ParCFD 2011, arXiv:1109.3524v1 [cs.CE] (16 Sep 2011)
@article{2011arXiv1109.3524L,
author={Layton}, S.~K and {Krishnan}, A. and {Barba}, L.~A.},
title={"{cuIBM — A GPU-accelerated Immersed Boundary Method}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1109.3524},
primaryClass={"cs.CE"},
keywords={Computer Science – Computational Engineering, Finance, and Science},
year={2011},
month={sep},
adsurl={http://adsabs.harvard.edu/abs/2011arXiv1109.3524L},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
A projection-based immersed boundary method is dominated by sparse linear algebra routines. Using the open-source Cusp library, we observe a speedup (with respect to a single CPU core) which reflects the constraints of a bandwidth-dominated problem on the GPU. Nevertheless, GPUs offer the capacity to solve large problems on commodity hardware. This work includes validation and a convergence study of the GPU-accelerated IBM, and various optimizations.
September 19, 2011 by hgpu
