On the Performance of an Algebraic Multigrid Solver on Multicore Clusters
Algebraic multigrid (AMG) solvers have proven to be extremely efficient on distributed-memory architectures. However, when executed on modern multicore cluster architectures, we face new challenges that can significantly harm AMG's performance. We discuss our experiences on such an architecture and present a set of techniques that help users to overcome the associated problems, including thread and process pinning and correct memory associations. We have implemented most of the techniques in a MultiCore SUPport library (MCSup), which helps to map OpenMP applications to multicore machines. We present results using both an MPI-only and a hybrid MPI/OpenMP model.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 986057
- Report Number(s):
- LLNL-CONF-420726
- Country of Publication:
- United States
- Language:
- English
Similar Records
On the Performance of an Algebraic MultigridSolver on Multicore Clusters
Challenges of Algebraic Multigrid across Multicore Architectures
Modeling the Performance of an Algebraic Multigrid Cycle Using Hybrid MPI/OpenMP
Conference
·
Thu Apr 29 00:00:00 EDT 2010
·
OSTI ID:1012429
Challenges of Algebraic Multigrid across Multicore Architectures
Conference
·
Mon Apr 12 00:00:00 EDT 2010
·
OSTI ID:1013213
Modeling the Performance of an Algebraic Multigrid Cycle Using Hybrid MPI/OpenMP
Conference
·
Sat Sep 01 00:00:00 EDT 2012
· 2012 41st International Conference on Parallel Processing; 10-13 Sept. 2012; Pittsburgh, PA, USA
·
OSTI ID:1567579