Skip to Main content Skip to Navigation

Computing least squares condition numbers on hybrid multicore/GPU systems

Marc Baboulin 1, 2, * Jack Dongarra 3 Rémi Lacroix 4
* Corresponding author
1 POSTALE - Performance Optimization by Software Transformation and Algorithms & Librairies Enhancement
LRI - Laboratoire de Recherche en Informatique, Inria Saclay - Ile de France
2 ParSys - LRI - Systèmes parallèles (LRI)
LRI - Laboratoire de Recherche en Informatique
4 ALPINES - Algorithms and parallel tools for integrated numerical simulations
LJLL - Laboratoire Jacques-Louis Lions, Inria Paris-Rocquencourt, INSMI - Institut National des Sciences Mathématiques et de leurs Interactions
Abstract : This paper presents an efficient computation for least squares conditioning or estimates of it. We propose performance results using new routines on top of the multicore-GPU library MAGMA. This set of routines is based on an efficient computation of the variance-covariance matrix for which, to our knowledge, there is no implementation in current public domain libraries LAPACK and ScaLAPACK.
Complete list of metadatas
Contributor : Marc Baboulin <>
Submitted on : Friday, February 14, 2014 - 6:53:52 PM
Last modification on : Thursday, March 26, 2020 - 9:25:00 PM
Document(s) archivé(s) le : Thursday, May 15, 2014 - 10:46:16 AM


Files produced by the author(s)


  • HAL Id : hal-00947204, version 1



Marc Baboulin, Jack Dongarra, Rémi Lacroix. Computing least squares condition numbers on hybrid multicore/GPU systems. [Research Report] RR-8479, 2014. ⟨hal-00947204v1⟩



Record views


Files downloads