Name
Abstract | ||
|---|---|---|
In this paper, we address the exploitation of data parallelism for the solution of sparse symmetric positive definite linear systems via iterative methods on Graphics Processing Units (GPUs). In particular, we accelerate the preconditioned CG-based iterative solver underlying the incomplete LU decomposition package (ILUPACK) by off-loading the most expensive computations i.e., The solution of sparse triangular systems and sparse matrix-vector products-to the hardware accelerator. The results collected using GPUs from the two most recent generations from NVIDIA (\"Fermi\" and \"Kepler\") and a benchmark test bed of sparse linear systems show that the GPU-enabled implementations deliver a notable reduction of the execution time, while maintaining the convergence rate and numerical properties of the original ILUPACK solver. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/ISPDC.2014.19 | ISPDC |
Keywords | Field | DocType |
sparse linear systems, iterative solvers, conjugate gradient method, incomplete lu factorization, gpu,linear systems,data parallelism,coprocessors,parallel processing,incomplete lu factorization,vectors,hardware accelerator,kepler,sparse matrices,iterative methods,kernel,convergence rate,fermi,acceleration,convergence | Conjugate gradient method,Computer science,Sparse approximation,Parallel computing,Data parallelism,Computational science,Incomplete LU factorization,Hardware acceleration,Solver,Stone method,LU decomposition | Conference |
ISSN | Citations | PageRank |
2379-5352 | 3 | 0.53 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| José Ignacio Aliaga | 1 | 75 | 15.18 |
| Matthias Bollhöfer | 2 | 186 | 17.75 |
| Ernesto Dufrechou | 3 | 25 | 11.02 |
| Pablo Ezzatti | 4 | 124 | 28.24 |
| Enrique S. Quintana-Ortí | 5 | 1317 | 150.59 |