1. D. Goudin. Mise en oe uvre d'une Bibliothèque d'Outils pour la Résolution Parallèle Hautes Performances par Méthode Directe de Grands Systèmes Linéaires Creux et application à un Code de Mécanique des Structures. PhD thesis, LaBRI, Université Bordeaux I, Talence, France, November 2000. Keyword(s): Sparse. Abstract:

   This thesis initially concerned the parallelization of the OSSAU software code from CEA/CESTA. The application domain of this software is the vectorized structural mechanics; the code itself is non linear in time and in two or three dimensions. This investigation leads to the conception and the implementation of a parallel high performance software processing chain for the assembly and the resolution of sparse linear systems by direct methods. The final objective is a validation of the OSSAU code for three dimension problems with several millions of unknowns.

   
   

   @PhdThesis{t:LaBRI::DG2k,
   author = "Goudin, D.",
   title = "{M}ise en \oe uvre d'une {B}iblioth\`eque d'{O}utils pour la {R}\'esolution {P}arall\`ele {H}autes {P}erformances par {M}\'ethode {D}irecte de {G}rands {S}yst\`emes {L}in\'eaires {C}reux et application \`a un {C}ode de {M}\'ecanique des {S}tructures",
   school = "LaBRI, Universit\'e Bordeaux I",
   year = "2000",
   OPTkey = {},
   address = {Talence, France},
   OPTtype = {},
   month = nov,
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/these_goudin.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {This thesis initially concerned the parallelization of the OSSAU software code from CEA/CESTA. The application domain of this software is the vectorized structural mechanics; the code itself is non linear in time and in two or three dimensions. This investigation leads to the conception and the implementation of a parallel high performance software processing chain for the assembly and the resolution of sparse linear systems by direct methods. The final objective is a validation of the OSSAU code for three dimension problems with several millions of unknowns.} 
   }
   




2. P. Ramet. Optimisation de la Communication et de la Distribution des Données pour des Solveurs Parallèles Directs en Algèbre Linéaire Dense et Creuse. PhD thesis, LaBRI, Université Bordeaux I, Talence, France, January 2000. Keyword(s): Overlap, Sparse. Abstract:

   This thesis deals with the high performance computation problems and more specifically with those of scientific parallel computation for irregular real-world applications. In the first part, we describe a method for overlapping communications on parallel computers with distributed memory. This method has resulted in a generic computation scheme for the optimal packet size. We also tackle the problem of finding the optimal computation grain for the Cholesky factorization algorithm for dense matrices. The goal of this study is to exploit the irregularity induced by the matrix symmetry. Based on this work we have developped a portable software library providing an efficient application context for these techniques. The second part of this thesis presents and analyses a general algorithm for the computation of an efficient static scheduling of block computations, developped especially for a parallel direct sparse linear factorization based on a combination of 1D and 2D block distributions. Our solver uses a supernodal Fan-In approach and is fully driven by our static scheduling algorithm. Compared to the existing parallel direct solvers our solver shows very favorable performance results.

   
   

   @PhdThesis{t:LaBRI::PR2k,
   author = "Ramet, P.",
   title = {{O}ptimisation de la {C}ommunication et de la {D}istribution des {D}onn\'ees pour des {S}olveurs Parall\`eles {D}irects en {A}lg\`ebre {L}in\'eaire {D}ense et {C}reuse},
   school = {LaBRI, Universit\'e Bordeaux I},
   year = "2000",
   OPTkey = {},
   address = {Talence, France},
   OPTtype = {},
   month = jan,
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/publications/these_ramet.pdf},
   KEYWORDS = "Overlap, Sparse",
   ABSTRACT = {This thesis deals with the high performance computation problems and more specifically with those of scientific parallel computation for irregular real-world applications. 
   
   In the first part, we describe a method for overlapping communications on parallel computers with distributed memory. This method has resulted in a generic computation scheme for the optimal packet size. We also tackle the problem of finding the optimal computation grain for the Cholesky factorization algorithm for dense matrices. The goal of this study is to exploit the irregularity induced by the matrix symmetry. Based on this work we have developped a portable software library providing an efficient application context for these techniques. 
   
   The second part of this thesis presents and analyses a general algorithm for the computation of an efficient static scheduling of block computations, developped especially for a parallel direct sparse linear factorization based on a combination of 1D and 2D block distributions. Our solver uses a supernodal Fan-In approach and is fully driven by our static scheduling algorithm. Compared to the existing parallel direct solvers our solver shows very favorable performance results.} 
   }
   

1. D. Goudin, P. Hénon, F. Pellegrini, P. Ramet, J. Roman, and J.-J. Pesque. Parallel Sparse Linear Algebra and Application to Structural Mechanics. Numerical Algorithms, 24:371-391, 2000. Keyword(s): Sparse. Abstract:

   The framework of this paper is the parallelization of a plasticity algorithm that uses an implicit method and an incremental approach. More precisely, we will focus on some specific parallel sparse linear algebra algorithms which are the most time-consuming steps to solve efficiently such an engineering application. First, we present a general algorithm which computes an efficient static scheduling of block computations for parallel sparse linear factorization. The associated solver, based on a supernodal fan-in approach, is fully driven by this scheduling. Second we describe a scalable parallel assembly algorithm based on a distribution of elements induced by the previous distribution for the blocks of the sparse matrix. We give an overview of these algorithms and present performance results on an IBM SP2 for a collection of grid and irregular problems.

   
   

   @Article{A:LaBRI::GHPRRP3,
   author = "Goudin, D. and H\'enon, P. and Pellegrini, F. and Ramet, P. and Roman, J. and Pesque, J.-J.",
   title = "{P}arallel Sparse Linear Algebra and Application to Structural Mechanics",
   journal = "{N}umerical {A}lgorithms",
   year = "2000",
   OPTkey = {},
   volume = {24},
   OPTnumber = {},
   pages = {371--391},
   OPTmonth = {},
   OPTannote = {},
   OPTURL = {http://www.labri.fr/~ramet/restricted/baltzer99.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {The framework of this paper is the parallelization of a plasticity algorithm that uses an implicit method and an incremental approach. More precisely, we will focus on some specific parallel sparse linear algebra algorithms which are the most time-consuming steps to solve efficiently such an engineering application. First, we present a general algorithm which computes an efficient static scheduling of block computations for parallel sparse linear factorization. The associated solver, based on a supernodal fan-in approach, is fully driven by this scheduling. Second we describe a scalable parallel assembly algorithm based on a distribution of elements induced by the previous distribution for the blocks of the sparse matrix. We give an overview of these algorithms and present performance results on an IBM SP2 for a collection of grid and irregular problems.} 
   }
   

1. D. Goudin. Assemblage parallèle d'une matrice et/ou d'un second membre: Application à la Parallélisation d'un Code de Mécanique des Structures. In ACTES RenPar'2000, Besancon, France, 2000. Keyword(s): Sparse. Abstract:

   Cet article d\'ecrit un algorithme d'assemblage parall\`ele pouvant \^etre utilis\'e lors de la phase de r\'esolution de grands syst\'emes lin\'eaires creux. La premi\`ere partie traite des probl\'emes pos\`es par un code de m\'ecanique des structures nomm\'e OSSAU d\'evelopp\'e au CEA. La deuxi\'eme partie est consacr\'ee \`a la description de notre algorithme et \`a son int\'egration dans la cha\^ine logicielle EMILIO, d\'evelopp\'ee au LaBRI sein du th\'eme ALiENor.

   
   

   @InProceedings{c:LaBRI::dg2ka,
   author = "Goudin, D.",
   title = "Assemblage parall\`ele d'une matrice et/ou d'un second membre: Application \`a la Parall\'elisation d'un Code de M\'ecanique des Structures",
   OPTcrossref = "",
   OPTkey = "",
   OPTeditor = "",
   OPTvolume = "",
   OPTnumber = "",
   OPTseries = "",
   pages = "",
   booktitle = "ACTES RenPar'2000",
   year = "2000",
   OPTorganization = "",
   OPTpublisher = "",
   address = "Besancon, France",
   OPTmonth = "",
   OPTnote = "",
   OPTannote = "",
   URL = {http://www.labri.fr/~ramet/restricted/renpar2k.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {Cet article d\'ecrit un algorithme d'assemblage parall\`ele pouvant \^etre utilis\'e lors de la phase de r\'esolution de grands syst\'emes lin\'eaires creux. La premi\`ere partie traite des probl\'emes pos\`es par un code de m\'ecanique des structures nomm\'e OSSAU d\'evelopp\'e au CEA. La deuxi\'eme partie est consacr\'ee \`a la description de notre algorithme et \`a son int\'egration dans la cha\^ine logicielle EMILIO, d\'evelopp\'ee au LaBRI sein du th\'eme ALiENor.} 
   }
   




2. D. Goudin, P. Hénon, F. Pellegrini, P. Ramet, and J. Roman. Résolution parallèle de grands systèmes linéaires creux. In Proceedings of JSFT'2000, Monastir, Tunisia, October 2000. Keyword(s): Sparse. Abstract:

   Cet article pr\'esente un tour d'horizon concernant les principes et techniques pouvant \^etre utilis\'ees pour la r\'esolution de grands syst\`emes lin\'eaires creux. Il pr\'esente en particulier les travaux r\'ealis\'es au LaBRI au sein du th\`eme ALiENor consacr\'es \`a la r\'esolution parall\`ele haute performance par m\'ethodes directes.

   
   

   @InProceedings{c:LaBRI::jsft,
   author = "Goudin, D. and H\'enon, P. and Pellegrini, F. and Ramet, P. and Roman, J.",
   title = "R\'esolution parall\`ele de grands syst\`emes lin\'eaires creux",
   booktitle = "Proceedings of {JSFT'2000}",
   OPTcrossref = {},
   OPTkey = {},
   OPTeditor = {},
   OPTvolume = {},
   OPTnumber = {},
   OPTseries = {},
   year = "2000",
   OPTorganization = {},
   OPTpublisher = {},
   address = {Monastir, Tunisia},
   month = oct,
   OPTpages = {},
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/jsft2k.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {Cet article pr\'esente un tour d'horizon concernant les principes et techniques pouvant \^etre utilis\'ees pour la r\'esolution de grands syst\`emes lin\'eaires creux. Il pr\'esente en particulier les travaux r\'ealis\'es au LaBRI au sein du th\`eme ALiENor consacr\'es \`a la r\'esolution parall\`ele haute performance par m\'ethodes directes.} 
   }
   




3. D. Goudin, P. Hénon, F. Pellegrini, P. Ramet, J. Roman, and J-J. Pesque. Algèbre Linéaire Creuse Hautes Performances : Application à la Mécanique des Structures. In iHPerf'2000, Aussois, France, December 2000. Keyword(s): Sparse.
   

   @InProceedings{c:LaBRI::ihperf_2k,
   author = "Goudin, D. and H\'enon, P. and Pellegrini, F. and Ramet, P. and Roman, J. and Pesque, J-J.",
   title = "{A}lg\`ebre {L}in\'eaire {C}reuse {H}autes {P}erformances : {A}pplication \`a la {M}\'ecanique des {S}tructures",
   booktitle = "{iHPerf'2000}",
   OPTcrossref = {},
   OPTkey = {},
   OPTeditor = {},
   OPTvolume = {},
   OPTnumber = {},
   OPTseries = {},
   year = "2000",
   OPTorganization = {},
   OPTpublisher = {},
   address = {Aussois, France},
   month = dec,
   OPTpages = {},
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/ihperf2k.ps},
   KEYWORDS = "Sparse" 
   }
   




4. D. Goudin, P. Hénon, F. Pellegrini, P. Ramet, J. Roman, and J.-J. Pesque. Description of the EMILIO Software Processing Chain and Application to Structural Mechanics. In Proceedings of PMAA'2K, Neuchatel, Swiss, August 2000. Keyword(s): Sparse.
   

   @InProceedings{C:LaBRI::PMAA_2K2,
   author = "Goudin, D. and H\'enon, P. and Pellegrini, F. and Ramet, P. and Roman, J. and Pesque, J.-J.",
   title = "Description of the EMILIO Software Processing Chain and Application to Structural Mechanics",
   booktitle = "Proceedings of {PMAA}'2K",
   OPTcrossref = {},
   OPTkey = {},
   OPTeditor = {},
   OPTvolume = {},
   OPTnumber = {},
   OPTseries = {},
   year = "2000",
   OPTorganization = {},
   OPTpublisher = {},
   address = {Neuchatel, Swiss},
   month = aug,
   OPTpages = {},
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/pmaa1.ps},
   KEYWORDS = "Sparse" 
   }
   




5. D. Goudin, P. Hénon, F. Pellegrini, P. Ramet, J. Roman, and J.-J. Pesque. Parallel Sparse Linear Algebra and Application to Structural Mechanics. In European ACTC Workshop, Paris, France, May 2000. Keyword(s): Sparse.
   

   @InProceedings{C:LaBRI::actc,
   author = "Goudin, D. and H\'enon, P. and Pellegrini, F. and Ramet, P. and Roman, J. and Pesque, J.-J.",
   title = "Parallel Sparse Linear Algebra and Application to Structural Mechanics",
   booktitle = "{European ACTC Workshop}",
   OPTcrossref = {},
   OPTkey = {},
   OPTeditor = {},
   OPTvolume = {},
   OPTnumber = {},
   OPTseries = {},
   year = {2000},
   OPTorganization = {},
   OPTpublisher = {},
   address = {Paris, France},
   month = may,
   OPTpages = {},
   OPTnote = {},
   OPTannote = {},
   KEYWORDS = "Sparse" 
   }
   




6. D. Goudin and J. Roman. A scalable parallel assembly for irregular meshes based on a block distribution for a parallel block direct solver. In Proceedings of PARA'2000, volume 1947 of LNCS, Bergen, Norway, 2000. Springer Verlag. Keyword(s): Sparse. Abstract:

   This paper describes a distribution of elements for irregular finite element meshes as well as the associated parallel assembly algorithm, in the context of parallel solving of the resulting sparse linear system using a direct block solver. These algorithms are integrated in the software processing chain EMILIO being developped at LaBRI for structural mechanics applications. Some illustrative numerical experiments on IBM SP2 validate this study.

   
   

   @InProceedings{C:LaBRI::gopara,
   author = "Goudin, D. and Roman, J.",
   title = "A scalable parallel assembly for irregular meshes based on a block distribution for a parallel block direct solver",
   year = "2000",
   booktitle = "Proceedings of {PARA'2000}",
   series = "LNCS",
   volume = "1947",
   address = "Bergen, Norway",
   publisher = "Springer Verlag",
   URL = {http://www.labri.fr/~ramet/restricted/para2k.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {This paper describes a distribution of elements for irregular finite element meshes as well as the associated parallel assembly algorithm, in the context of parallel solving of the resulting sparse linear system using a direct block solver. These algorithms are integrated in the software processing chain EMILIO being developped at LaBRI for structural mechanics applications. Some illustrative numerical experiments on IBM SP2 validate this study.} 
   }
   




7. P. Hénon, P. Ramet, and J. Roman. PaStiX: A High-Performance Parallel Direct Solver for Sparse Symmetric Definite Systems. In Proceedings of PMAA'2K, Neuchatel, Swiss, August 2000. Keyword(s): Sparse.
   

   @InProceedings{C:LaBRI::PMAA_2K1,
   author = "H\'enon, P. and Ramet, P. and Roman, J.",
   title = "PaStiX: A High-Performance Parallel Direct Solver for Sparse Symmetric Definite Systems",
   booktitle = "Proceedings of {PMAA}'2K",
   OPTcrossref = {},
   OPTkey = {},
   OPTeditor = {},
   OPTvolume = {},
   OPTnumber = {},
   OPTseries = {},
   year = "2000",
   OPTorganization = {},
   OPTpublisher = {},
   address = {Neuchatel, Swiss},
   month = aug,
   OPTpages = {},
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/pmaa2.ps},
   KEYWORDS = "Sparse" 
   }
   




8. P. Hénon, P. Ramet, and J. Roman. PaStiX: A Parallel Sparse Direct Solver Based on a Static Scheduling for Mixed 1D/2D Block Distributions. In Proceedings of Irregular'2000 workshop of IPDPS, volume 1800 of LNCS, Cancun, Mexico, pages 519-525, May 2000. Springer Verlag. Keyword(s): Sparse. Abstract:

   We present and analyze a general algorithm which computes an efficient static scheduling of block computations for a parallel $L.D.L^{t}$ factorization of sparse symmetric positive definite systems based on a combination of 1D and 2D block distributions. Our solver uses a supernodal fan-in approach and is fully driven by this scheduling. We give an overview of the algorithm and present performance results and comparisons with PSPASES on an IBM-SP2 with 120 MHz Power2SC nodes for a collection of irregular problems.

   
   

   @InProceedings{C:LaBRI::hrr2k,
   author = {H\'enon, P. and Ramet, P. and Roman, J.},
   title = {{PaStiX}: {A} {P}arallel {S}parse {D}irect {S}olver {B}ased on a {S}tatic {S}cheduling for {M}ixed {1D/2D} {B}lock {D}istributions},
   booktitle = {Proceedings of {I}rregular'2000 workshop of IPDPS},
   OPTcrossref = {},
   OPTkey = {},
   pages = {519-525},
   year = 2000,
   OPTeditor = {},
   OPTnumber = {},
   volume = 1800,
   series = {LNCS},
   address = {Cancun, Mexico},
   month = may,
   OPTorganization = {},
   publisher = {Springer Verlag},
   OPTnote = {},
   OPTannote = {},
   URL = {http://www.labri.fr/~ramet/restricted/irregular2K.ps},
   KEYWORDS = "Sparse",
   ABSTRACT = {We present and analyze a general algorithm which computes an efficient static scheduling of block computations for a parallel $L.D.L^{t}$ factorization of sparse symmetric positive definite systems based on a combination of 1D and 2D block distributions. Our solver uses a supernodal fan-in approach and is fully driven by this scheduling. We give an overview of the algorithm and present performance results and comparisons with PSPASES on an IBM-SP2 with 120 MHz Power2SC nodes for a collection of irregular problems.} 
   }
   

1. D. Goudin. Mise en oeuvre d'une Bibliothèque d'Outils pour la Résolution par Méthode Directe de Grands Systèmes Linéaires Creux Symétriques Définis Positifs sur Machine Parallèle. Technical report, C.E.A. / C.E.S.T.A, 2000. Note: Rapport Final de la Deuxième Partie. Keyword(s): Sparse.
   

   @TechReport{f:LaBRI::go2ka,
   author = "Goudin, D.",
   title = "Mise en oeuvre d'une Biblioth\`eque d'Outils pour la R\'esolution par M\'ethode Directe de Grands Syst\`emes Lin\'eaires Creux Sym\'etriques D\'efinis Positifs sur Machine Parall\`ele",
   institution = "C.E.A. / C.E.S.T.A",
   year = "2000",
   note = "Rapport Final de la Deuxi\`eme Partie",
   KEYWORDS = "Sparse" 
   }
   

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All person copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Les documents contenus dans ces répertoires sont rendus disponibles par les auteurs qui y ont contribué en vue d'assurer la diffusion à temps de travaux savants et techniques sur une base non-commerciale. Les droits de copie et autres droits sont gardés par les auteurs et par les détenteurs du copyright, en dépit du fait qu'ils présentent ici leurs travaux sous forme électronique. Les personnes copiant ces informations doivent adhérer aux termes et contraintes couverts par le copyright de chaque auteur. Ces travaux ne peuvent pas être rendus disponibles ailleurs sans la permission explicite du détenteur du copyright.

This document was translated from BibT_EX by bibtex2html