Title :
Task-Based Parallelization of Unstructured Meshes Assembly Using D&C Strategy
Author :
Petit, Eric ; Thebault, Loic ; Moller, Nathalie ; Jalby, William ; Quang Dinh
Author_Institution :
PRISM, Univ. of Versailles UVSQ, Versailles, France
Abstract :
This paper introduces our work on parallelizing unstructured mesh applications in shared memory. Our approach uses task-based runtime and Divide and Conquer recursive bissection strategy for the work decomposition. We demonstrate this parallelization strategy on the unstructured mesh assembly step of an industrial CFD application. Preliminary results show a speedup of 1.8x on a 12 cores node and a performance on Xeon Phi equivalent to 15 Ivy Bridge cores, with a parallel efficiency higher than 90% on the physical cores.
Keywords :
computational fluid dynamics; divide and conquer methods; mesh generation; multi-threading; parallel algorithms; shared memory systems; D&C strategy; Ivy Bridge cores; Xeon Phi; cores node; divide-and-conquer recursive bissection strategy; industrial CFD application; multithreading; parallel algorithms; parallel programming; shared memory; task-based parallelization strategy; task-based runtime; unstructured mesh assembly step; work decomposition; Assembly; Concurrent computing; Finite element analysis; Image color analysis; Parallel processing; Runtime; Synchronization; CFD; Cilk++; MPI; Multithreading; Parallel algorithms; Parallel programming; mesh;
Conference_Titel :
High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS), 2014 IEEE Intl Conf on
Print_ISBN :
978-1-4799-6122-1
DOI :
10.1109/HPCC.2014.150
