Exploiting thread-level parallelism in the iterative solution of sparse linear systems

Authors:
José I. Aliaga;Matthias Bollhöfer;Alberto F. Martın;Enrique S. Quintana-Ortı
Affiliations:
Dpto. de Ingenierıa y Ciencia de Computadores, Universidad Jaime I, 12.071-Castellón, Spain;Institut Computational Mathematics, TU Braunschweig, 38106 Braunschweig, Germany;Dpto. de Ingenierıa y Ciencia de Computadores, Universidad Jaime I, 12.071-Castellón, Spain;Dpto. de Ingenierıa y Ciencia de Computadores, Universidad Jaime I, 12.071-Castellón, Spain
Venue:
Parallel Computing
Year:
2011

Citing 14
Cited 3

The approximate Dirichlet domain decomposition method. Part I: an algebraic approach

Computing
Scalable iterative solution of sparse linear systems

Parallel Computing
A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs

SIAM Journal on Scientific Computing
Parallel threshold-based ILU factorization

SC '97 Proceedings of the 1997 ACM/IEEE conference on Supercomputing
A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling

SIAM Journal on Matrix Analysis and Applications
Two-level dynamic scheduling in PARDISO: improved scalability on shared memory multiprocessing systems

Parallel Computing - Parallel matrix algorithms and applications
Hybridizing Nested Dissection and Halo Approximate Minimum Degree for Efficient Sparse Matrix Ordering

Proceedings of the 11 IPPS/SPDP'99 Workshops Held in Conjunction with the 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing
Iterative Methods for Sparse Linear Systems

Iterative Methods for Sparse Linear Systems
Solving unsymmetric sparse systems of linear equations with PARDISO

Future Generation Computer Systems - Special issue: Selected numerical algorithms
Multilevel Preconditioners Constructed From Inverse-Based ILUs

SIAM Journal on Scientific Computing
On Large-Scale Diagonalization Techniques for the Anderson Model of Localization

SIAM Review
Design, Tuning and Evaluation of Parallel Multilevel ILU Preconditioners

High Performance Computing for Computational Science - VECPAR 2008
Inertia-Revealing Preconditioning For Large-Scale Nonconvex Constrained Optimization

SIAM Journal on Scientific Computing
Algebraic Multilevel Preconditioner for the Helmholtz Equation in Heterogeneous Media

SIAM Journal on Scientific Computing

Parallelization of multilevel ILU preconditioners on distributed-memory multiprocessors

PARA'10 Proceedings of the 10th international conference on Applied Parallel and Scientific Computing - Volume Part I
Leveraging task-parallelism in energy-efficient ILU preconditioners

ICT-GLOW'12 Proceedings of the Second international conference on ICT as Key Technology against Global Warming
Benefits of using parallelized non-progressive network coding

Journal of Network and Computer Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

We investigate the efficient iterative solution of large-scale sparse linear systems on shared-memory multiprocessors. Our parallel approach is based on a multilevel ILU preconditioner which preserves the mathematical semantics of the sequential method in ILUPACK. We exploit the parallelism exposed by the task tree corresponding to the nested dissection hierarchy (task parallelism), employ dynamic scheduling of tasks to processors to improve load balance, and formulate all stages of the parallel PCG method conformal with the computation of the preconditioner to increase data reuse. Results on a CC-NUMA platform with 16 processors reveal the parallel efficiency of this solution.