论文信息 - Parallelization of Multilevel ILU Preconditioners on Distributed-Memory Multiprocessors

Parallelization of Multilevel ILU Preconditioners on Distributed-Memory Multiprocessors

In this paper we investigate the parallelization of the ILUPACK library for the solution of sparse linear systems on distributed-memory multiprocessors. The parallelization approach employs multilevel graph partitioning algorithms in order to identify a set of concurrent tasks and their dependencies, which are then statically mapped to processors. Experimental results on a cluster of Intel QuadCore processors report remarkable speed-ups.

[1] Enrique S. Quintana-Ortí,et al. Exploiting thread-level parallelism in the iterative solution of sparse linear systems , 2011, Parallel Comput..

[2] Marcus J. Grote,et al. Algebraic Multilevel Preconditioner for the Helmholtz Equation in Heterogeneous Media , 2009, SIAM J. Sci. Comput..

[3] Yousef Saad,et al. Multilevel Preconditioners Constructed From Inverse-Based ILUs , 2005, SIAM J. Sci. Comput..

[4] Vipin Kumar,et al. A Parallel Algorithm for Multilevel Graph Partitioning and Sparse Matrix Ordering , 1998, J. Parallel Distributed Comput..

[5] Vipin Kumar,et al. A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs , 1998, SIAM J. Sci. Comput..

[6] Enrique S. Quintana-Ortí,et al. Design, Tuning and Evaluation of Parallel Multilevel ILU Preconditioners , 2008, VECPAR.

[7] Timothy A. Davis,et al. Direct methods for sparse linear systems , 2006, Fundamentals of algorithms.

[8] Christian Bischof,et al. Parallel computing : architectures, algorithms and applications , 2008 .

[9] Enrique S. Quintana-Ortí,et al. Parallelization of Multilevel Preconditioners Constructed from Inverse-Based ILUs on Shared-Memory Multiprocessors , 2007, PARCO.

[10] Yousef Saad,et al. Iterative methods for sparse linear systems , 2003 .

[11] François Pellegrini,et al. PT-Scotch: A tool for efficient parallel graph ordering , 2008, Parallel Comput..