Abstract | ||
---|---|---|
This paper presents an investigation into the use of various mechanisms for improving the resilience of the fine-grained parallel algorithm for computing an incomplete LU factorization. These include various approaches to checkpointing as well as a study into the feasibility of using a self-stabilizing periodic correction step. Results concerning convergence of all of the self-stabilizing variants of the algorithm with respect to the occurrence of faults, and the impact of any sub-optimality in the produced incomplete L and U factors in Krylov subspace solvers are given. Numerical tests show that the simple algorithmic changes suggested here can ensure convergence of the fine-grained parallel incomplete factorization, and improve the performance of the resulting factors as preconditioners in Krylov subspace solvers in the presence of transient soft faults. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1016/j.suscom.2018.01.003 | Sustainable Computing: Informatics and Systems |
Keywords | Field | DocType |
Fault tolerance,Parallel preconditioning,Incomplete factorization,Asynchronous iterative methods,Self-stabilizing iterative algorithms | Convergence (routing),Krylov subspace,Numerical tests,Parallel algorithm,Computer science,Algorithm,Incomplete LU factorization,Factorization,Periodic graph (geometry) | Journal |
Volume | ISSN | Citations |
19 | 2210-5379 | 0 |
PageRank | References | Authors |
0.34 | 21 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Evan Coleman | 1 | 1 | 2.72 |
Evan Coleman | 2 | 1 | 2.72 |
Masha Sosonkina | 3 | 272 | 45.62 |