WorldCat Identities

Desprez, Frédéric

Overview
Works: 30 works in 49 publications in 2 languages and 642 library holdings
Genres: Conference proceedings 
Roles: Editor
Classifications: QA76.9.C58, 004.36
Publication Timeline
Key
Publications about  Frédéric Desprez Publications about Frédéric Desprez
Publications by  Frédéric Desprez Publications by Frédéric Desprez
Most widely held works by Frédéric Desprez
Grids, P2P and services computing by Frédéric Desprez ( )
10 editions published in 2010 in English and held by 375 WorldCat member libraries worldwide
& Quot;Grids, P2P and Services Computing", the 12th volume of the "CoreGRID" series, is based on the CoreGrid ERCIM Working Group Workshop on Grids, P2P and Service Computing in Conjunction with EuroPar 2009. The workshop will take place August 24th, 2009 in Delft, The Netherlands. Grids, P2P and Services Computing, an edited volume contributed by well-established researchers worldwide, will focus on solving research challenges for Grid and P2P technologies. Topics of interest include: Service Level Agreement, Data & Knowledge Management, Scheduling, Trust and Security, Network M
Initiation au parallélisme : concepts, architectures et algorithmes by Marc Gengler ( Book )
4 editions published between 1995 and 1996 in French and held by 91 WorldCat member libraries worldwide
Algorithms and tools for parallel computing on heterogeneous clusters ( Book )
4 editions published between 2006 and 2007 in English and held by 80 WorldCat member libraries worldwide
UNIX pour l'impatient / Julien Cassaigne et Frédéric Desprez by Paul W Abrahams ( Book )
1 edition published in 1999 in French and held by 38 WorldCat member libraries worldwide
Euro-Par 2012: Parallel processing workshops BDMC, CGWS, HeteroPar, HiBB, OMHI, Paraphrase, PROPER, Resilience, UCHPC, VHPC, Rhodes Islands, Greece, August 27-31, 2012. Revised selected papers by Ioannis Caragiannis ( )
2 editions published in 2013 in English and held by 16 WorldCat member libraries worldwide
This book constitutes thoroughly refereed post-conference proceedings of the workshops of the 18th International Conference on Parallel Computing, Euro-Par 2012, held in Rhodes Islands, Greece, in August 2012. The papers of these 10 workshops BDMC, CGWS, HeteroPar, HiBB, OMHI, Paraphrase, PROPER, UCHPC, VHPC focus on promotion and advancement of all aspects of parallel and distributed computing
Proceedings of the 6th international workshop on Virtualization Technologies in Distributed Computing Date by Frédéric Desprez ( )
1 edition published in 2012 in English and held by 7 WorldCat member libraries worldwide
Procédures de base pour le calcul scientifique sur machines parallèles à mémoire distrubuée by Frédéric Desprez ( Book )
2 editions published between 1994 and 2008 in French and held by 3 WorldCat member libraries worldwide
Le but de cette thèse est l'étude et l'implémentation de routines de base pour aider l'utilisateur de machines parallèles à mémoire distribuée à obtenir les meilleures performances avec un coût de développement moindre. Trois ensembles de routines sont étudiés. Le premier concerne des routines de communication sur réseau réconfigurable. La seconde bibliothèque fournit à l'utilisateur des routines portables de recouvrements calculs/communications transparents. Enfin, le dernier ensemble concerne des routines de calcul comme le produit de matrices, la factorisation LU et la transformée de Fourier bidimensionnelle. Une attention toute particulière est portée aux recouvrements calculs/communications. Enfin, une application des principes présentés tout au long de la thèse est donnée. Elle concerne la simulation d'un front de combustion
Proceedings of the 7th international workshop on Virtualization technologies in distributed computing by Frédéric Desprez ( )
1 edition published in 2013 in English and held by 2 WorldCat member libraries worldwide
Recouvrement des communications et des calculs, du matériel au logiciel by Frédérique Silber-Chaussumier ( Book )
1 edition published in 2002 in French and held by 1 WorldCat member library worldwide
Le but de cette thèse est l'étude des recouvrements des communications sur des machines parallèles à mémoire distribuée. Les recouvrements des communications sont des techniques très utilisées dans le but d'accélérer les exécutions de programmes parallèles. Notre première contribution est l'étude détaillée d'une implantation efficace de la bibliothèque de communication MPI sur une grappe de PC interconnectée par un réseau rapide. Cette étude montre tout d'abord que le matériel permet de recouvrir le transfert des données sur le réseau. Dans un deuxième temps, nous montrons pourquoi le logiciel de communication inter-processus, MPI, empêche le recouvrement des communications par le calcul. La sémantique des communications et l'implication du processeur de calcul dans la communication en sont les principales causes.Notre deuxième contribution est la validation des équations de gain d'un pipeline modélisées par Zory, par l'intermédiaire de l'application Sweep3D. Cette application nous a permis de montrer que sur des données de grande taille, l'utilisation de pipelines logiciels sur P processeurs peut apporter un gain en temps d'execution s'approchant d'un facteur P. Notre troisième contribution est l'évaluation a priori de la meilleurs configuration de l'espace des processeurs, c'est-à-dire mono, bi ou tridimensionnel. En établissant les équations des temps d'exécution d'une application par vague pour ces trois types de distribution en fonction des paramètres de la machine cible et de l'application, nous pouvons déterminer la distribution qui donnera le plus petit temps d'exécution
LAPACK working note 67 : performance complexity of LU factorization with efficient pipelining and overlap on a multiprocessor by Frédéric Desprez ( Book )
1 edition published in 1993 in English and held by 1 WorldCat member library worldwide
Abstract: "In this paper, we make efficient use of pipelining on LU decomposition with pivoting and a column-scattered data decomposition to derive precise variations of the computational complexities. We then compare these results with experiments on the Intel iPSC/860 and Paragon machines."
Automatic Deployment for Application Service Provider Environments by Pushpinder Kaur Chouhan ( Book )
1 edition published in 2006 in English and held by 1 WorldCat member library worldwide
The objective of the thesis is to improve the performance of a NES so as to use these environments efficiently. The very first problem is related to the applications scheduling on the selected servers. We have proposed and experimentally proved that the deadline scheduling with priority along with fallback mechanism can increase the efficiency of a NES. Another important factor that influences the efficiency of the NES environments is the mapping style of the environment's components on the available resources. We have shown theoretically that the optimal deployment on cluster is a Complete Spanning d-ary tree. Considering heterogeneous resources we presented a deployment heuristic, as finding the best deployment among heterogeneous resources is NP-complete. Finally, we gave a mathematical model that can analyze an existing deployment and can improve the performance of the deployment by finding and then removing the bottlenecks. Presented algorithms and heuristics are validated by implementing them to DIET, on different sites of Grid'5000
Étude et optimisation du comportement mémoire dans les méthodes parallèles de factorisation de matrices creuses by Abdou Guermouche ( Book )
1 edition published in 2004 in French and held by 1 WorldCat member library worldwide
Les méthodes directes de résolution de systèmes linéaires creux sont connues pour leurs besoins mémoire importants qui peuvent constituer une barrière au traitement de problèmes de grandes taille. De ce fait, les travaux effectués durant cette thèse ont porté d'une part sur l'étude du comportement mémoire d'un algorithme de factorisation de matrices creuses, en l'occurrence la méthode multifrontale, et d'autre part sur l'optimisation et la minimisation de la mémoire nécessaire au bon déroulement de la factorisation aussi bien dans un cadre séquentiel que parallèle. Ainsi, des algorithmes optimaux pour la minimisation de la mémoire ont été proposés pour le cas séquentiel. Pour le cas parallèle, nous avons introduit dans un premier temps des stratégies d'ordonnancement visant une amélioration du comportement mémoire de la méthode. Puis, nous les avons étendues pour avoir un objectif de performance tout en gardant un bon comportement mémoire. Enfin, dans le cas où l'ensemble des données à traiter a encore une taille plus importante que celle de la mémoire, il est nécessaire de concevoir des approches de factorisation out-of-core. Pour être efficaces, ces méthodes nécessitent d'une part de recouvrir les opérations d'entrées/sorties par des calculs, et d'autre part de réutiliser des données déjà présentes en mémoire pour réduire le volume d'entrées/sorties. Ainsi, une partie des travaux présentés dans cette thèse ont porté sur la conception de techniques out-of-core implicites adaptées au schéma des accès de la méthode multifrontale et reposant sur une modification de la politique de pagination du système d'exploitation à l'aide d'un outil bas-niveau (MMUM&MMUSSEL)
Gestion des bases de données biologiques sur grilles de calcul by Gaël Le Mahec ( Book )
1 edition published in 2008 in French and held by 1 WorldCat member library worldwide
L'utilisation de la grille pour la recherche en biologie et bioinformatique est un atout considérable, cependant de nouvelles problématiques apparaissent quant à la gestion des données ainsi que dans l'ordonnancement des tâches qui doit prendre en compte la taille et la disponibilité des données. Cette thèse aborde ces problématiques nouvelles en prenant en compte les spécificités des bases de données biologiques pour une utilisation efficace de la grille. Nous montrons l'intérêt des approches semi-statiques joignant réplications de données et ordonnancement des tâches. La mise en place d'une plateforme de soumission de requêtes BLAST a mené à la conception d'un nouveau gestionnaire de données pour l'intergiciel DIET : DAGDA. Au delà des applications de bioinformatique, ce gestionnaire de données peut répondre aux besoins de nombreuses applications portées sur les grilles de calcul
Découverte automatique des caractéristiques et capacités d'une plate-forme de calcul distribué by Martin Quinson ( Book )
1 edition published in 2003 in French and held by 1 WorldCat member library worldwide
This thesis is devoted to the monitoring of modern computational platforms in order to obtain relevant, up to date and accurate information about them. Often called Grids, those environments differ from the preceding parallel machines by their intrinsic heterogeneity and high dynamicity. This document is organized in three parts. The first one presents the specific difficulties introduced by this platform, highlighting them in a selection of grid infrastructure projects and detailing the existing solutions. The second part shows how to get efficiently quantitative informations about the grid capacities and their suitability to the needs of the routines to schedule. After a discussion of the problems encountered, we detail our approach which we call macro-benchmarking. We then present FAST, a tool implementing this methodology. We eventually detail how FAST is used in several other projects. The third part introduces how to get a more qualitative view of the grid characteristics such as the topology of the network interconnecting the hosts. After a study of the existing solutions in this domain, we present ALNeM our solution to automatically map the network without relying on specific execution privileges on the platform. This tool is based on GRAS, our framework for the development of grid infrastructure
Successive broadcasts on hypercube by Frédéric Desprez ( Book )
1 edition published in 1993 in English and held by 1 WorldCat member library worldwide
Abstract: "Broadcasting is an information dissemination problem in which information originating at one node of a communication network must be transmitted to all the other nodes as quickly as possible [1,2]. In this paper, we consider the problem in which all the nodes of a network must, by turns, broadcast a distinct message. We call this problem the successive broadcasts problem. Successive broadcasts is a communication pattern that appears in several parallel implementations of linear algebra algorithms on distributed memory multicomputers. Note that the successive broadcasts problem is different from the gossip problem [2] in which all the nodes must perform a broadcast in any order, even simultaneously. We present an algorithm solving the successive broadcasts problem on hypercubes. We derive a lower bound on the time of any successive broadcasts [sic] algorithms that shows that our algorithm is within a factor of 2 of the optimality."
La programmation d'applications scientifiques sur les ordinateurs parallèles à mémoire distribuée/ Frédéric Desprez et Bernard Tourancheau by Frédéric Desprez ( Book )
1 edition published in 1994 in French and held by 1 WorldCat member library worldwide
Scheduling and Dynamic Management of Applications over Grids by Ghislain Charrier ( Book )
1 edition published in 2010 in English and held by 1 WorldCat member library worldwide
The work presented in this Thesis is about scheduling applications in computational Grids. We study how to better manage jobs in a grid middleware in order to improve the performance of the platform. Our solutions are designed to work at the middleware layer, thus allowing to keep the underlying architecture unmodified. First, we propose a reallocation mechanism to dynamically tackle errors that occur during the scheduling. Indeed, it is often necessary to provide a runtime estimation when submitting on a parallel computer so that it can compute a schedule. However, estimations are inherently inaccurate and scheduling decisions are based on incorrect data, and are therefore wrong. The reallocation mechanism we propose tackles this problem by moving waiting jobs between several parallel machines in order to reduce the scheduling errors due to inaccurate runtime estimates. Our second interest in the Thesis is the study of the scheduling of a climatology application on the Grid. To provide the best possible performances, we modeled the application as a Directed Acyclic Graph (DAG) and then proposed specific scheduling heuristics. To execute the application on the Grid, the middleware uses the knowledge of the application to find the best schedule
 
moreShow More Titles
fewerShow Fewer Titles
Audience Level
0
Audience Level
1
  Kids General Special  
Audience level: 0.66 (from 0.00 for Procédure ... to 0.92 for Proceeding ...)
Languages
English (26)
French (12)
Covers