2026-04-14T07:57:07Zhttps://recercat.cat/oai/request

oai:recercat.cat:2117/182592026-02-07T09:08:38Zcom_2072_1033col_2072_452950

Design, tuning and evaluation of parallel multilevel ILU preconditioners Aliaga, José Ignacio Bollhoefer, Matthias Martín Huertas, Alberto Francisco Quintana Ortí, Enrique Salvador Universitat Politècnica de Catalunya. Departament de Resistència de Materials i Estructures a l'Enginyeria Àrees temàtiques de la UPC::Informàtica Parallel and distributing computing Sparse linear system Incomplete LU factorization In this paper, we present a parallel multilevel ILU preconditioner implemented with OpenMP. We employ METIS partitioning algorithms to decompose the computation into concurrent tasks, which are then scheduled to threads. Concretely, we combine decompositions which obtain significantly more tasks than processors, and the use of dynamic scheduling strategies in order to reduce the thread’s idle time, which it is shown to be the main source of overhead in our parallel algorithm. Experimental results on a shared-memory platform consisting of 16 processors report remarkable performance for our approach. Peer Reviewed Postprint (published version) 2008 Conference lecture Aliaga, J. I. [et al.]. Design, tuning and evaluation of parallel multilevel ILU preconditioners. A: International Meeting High Performance Computing for Computational Science. "High Performance Computing for Computational Science: VECPAR 2008: 8th international conference: Toulouse, France, June 24 -27, 2008: revised selected papers". Toulouse: 2008, p. 314-327. 978-3-540-92858-4 https://hdl.handle.net/2117/18259 10.1007/978-3-540-92859-1_28 eng http://link.springer.com/chapter/10.1007/978-3-540-92859-1_28?null Restricted access - publisher's policy 14 p. application/pdf