caa a22 4500 463202996 CHVBK 20180405153117.0 cr unu---uuuuu 170326e20070501xx s 000 0 eng 10.1007/s00200-007-0036-y doi (NATIONALLICENCE)springer-10.1007/s00200-007-0036-y Towards an accurate performance modeling of parallel sparse factorization [Elektronische Daten] [Laura Grigori, Xiaoye Li] We present a simulation-based performance model to analyze a parallel sparse LU factorization algorithm on modern cached-based, high-end parallel architectures. We consider supernodal right-looking parallel factorization on a bi-dimensional grid of processors, that uses static pivoting. Our model characterizes the algorithmic behavior by taking into account the underlying processor speed, memory system performance, as well as the interconnect speed. The model is validated using the implementation in the SuperLU_DIST linear system solver, the sparse matrices from real application, and an IBM POWER3 parallel machine. Our modeling methodology can be adapted to study performance of other types of sparse factorizations, such as Cholesky or QR, and on different parallel machines. Springer-Verlag, 2007 Parallel sparse factorization nationallicence Performance modeling nationallicence Distributed parallel machine nationallicence Grigori Laura INRIA Futurs, Parc Club Orsay Universite, 91893, Orsay, France aut Li Xiaoye Lawrence Berkeley National Laboratory, One Cyclotron Road, 94720, Berkeley, CA, USA aut Applicable Algebra in Engineering, Communication and Computing Springer-Verlag 18/3(2007-05-01), 241-261 0938-1279 18:3<241 2007 18 200 https://doi.org/10.1007/s00200-007-0036-y text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/s00200-007-0036-y text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Grigori Laura INRIA Futurs, Parc Club Orsay Universite, 91893, Orsay, France aut NATIONALLICENCE 700 1- Li Xiaoye Lawrence Berkeley National Laboratory, One Cyclotron Road, 94720, Berkeley, CA, USA aut NATIONALLICENCE 773 0- Applicable Algebra in Engineering, Communication and Computing Springer-Verlag 18/3(2007-05-01), 241-261 0938-1279 18:3<241 2007 18 200 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer