Fine-grain locality-aware parallel scheme for anisotropic mesh adaptation
Rakotoarivelo, Hoby, Franck Ledoux, Franck Pommereau
Proceedings, 25th International Meshing Roundtable, Elsevier, Science Direct, September 26-30 2016
25th International Meshing Roundtable
Washington DC, U.S.A.
September 26-30, 2016
Hoby Rakotoarivelo, CEA, FR, firstname.lastname@example.org
Franck Ledoux, CEA, FR, email@example.com
Franck Pommereau, Laboratoire IBISC, Université Paris-Saclay, FR, firstname.lastname@example.org
In this paper, we provide a fine-grained parallel scheme for anisotropic mesh adaptation on NUMA architectures.
Data dependencies are expressed by a graph for each kernel, and concurrency is extracted through fine-grained graph coloring. Tasks are structured into bulk-synchronous steps to avoid data races and to aggregate shared-data accesses.
To ensure performance prediction, time cost and load imbalance are theoretically characterized.
The devised scheme was evaluated on a 4 NUMA node (2-socket) machine, and a mean efficiency of 70% was reached on 32 cores for 3 kernels out of 4. The impact of irregular degree distribution and data layout on scalability is highlighted.
Download Full Paper (PDF)
Contact author(s) or publisher for availability and copyright information on above referenced article