20th International Meshing Roundtable
October 23-26, 2011
Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, U.S.A.
Email: firstname.lastname@example.org, email@example.com
We present a GPU-based algorithm for optimizing the shape of
elements in tetrahedral volume meshes. To exploit the massive parallelism
available in modern GPU hardware, optimization occurs on a per-vertex basis
using only local neighborhood information. A classical derivative-free numerical
optimization method is employed to optimize the minimum dihedral angle
of the component tetrahedron elements. Preliminary results demonstrate the
GPU as a promising platform for mesh optimization with notable speedups.
The final mesh exhibits high-quality comparable to state-of-the-art CPU techniques.
Ongoing efforts are focused on further performance optimizations as
well as scaling to multiple GPUs.
Download Full Paper (PDF Format)
Contact author(s) or publisher for availability and copyright information on above referenced article