Mesh adaptation for moving objects on shared memory hardware
Ibanez, Dan, Mark Shephard
Research Note, 25th International Meshing Roundtable, Sandia National Laboratories, September 26-30 2016
25th International Meshing Roundtable
Washington DC, U.S.A.
September 26-30, 2016
Dan Ibanez, Rensselaer Polytechnic Institute, US, firstname.lastname@example.org
Mark Shephard, Rensselaer Polytechnic Institute, US, email@example.com
Research Note Abstract
We present a general algorithm for scheduling the application of cavity-based mesh modifications which enables parallel mesh adaptation on shared-memory parallel hardware including accelerators such as GPUs. We demonstrate this method on a mesh with objects moving in the fluid, and performance is studied on hardware including an Intel Knights Landing processor and an NVidia K80 GPU. Our code ports seamlessly between these architectures and has no architecture-specific algorithms.
Download Full Paper (PDF)
Contact author(s) or publisher for availability and copyright information on above referenced article