On the Performance of an Algebraic Multigrid Solver on Multicore Clusters
Allison Baker (Lawrence Livermore National Laboratory)
Martin Schulz (Lawrence Livermore National Laboratory)
Ulrike Yang (Lawrence Livermore National Laboratory)
Abstract:
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.
Keywords:
Parallel and Distributed Computing, Numerical Algorithms for CS&E, Performance Analysis