carrier image

GPU-Based Parallel Algorithms for Delaunay Mesh Refinement

Pande, Sushrut, Supratim Biswas, Amitava De

24th International Meshing Roundtable, Elsevier Ltd., pp.Research Note, October 12-14 2015


24th International Meshing Roundtable
Austin, TX
October 12-14,2014

University of California, Berkeley, USA.
Indian Institute of Technology (IIT) Bombay, India.

This research note reports the design and implementation of a novel, robust parallel algorithm using GPUs (graphics processing units) to generate triangular meshes over complex, unstructured domains or with an unsuitable initial discretization. Algorithms have been implemented using the CUDA architecture and compared with their serial counterparts for several benchmark case studies. Results demonstrate that the GPU-based parallel algorithms developed are several orders of magnitude faster than their serial equivalents.

Download Full Paper (PDF Format)

Contact author(s) or publisher for availability and copyright information on above referenced article