Arnaud Legrand
59
Documents
Présentation
### Contact:
arnaud.legrand@imag.fr
### Webpage:
[Academic Page ](http://mescal.imag.fr/membres/arnaud.legrand/)
### Resume:
Arnaud Legrand is the leader of the [POLARIS](https://team.inria.fr/polaris/) team. He is a [CNRS](http://www.cnrs.fr/) research scientist at the [LIG](http://www.liglab.fr/). His research targets the management (mostly from an algorithmic point of view, i.e., scheduling, load balancing, fairness, game theory….) and performance evaluation (in particular through simulation, visualization, statistical analysis, …) of large scale distributed computing infrastructures such as clusters, grids, desktop grids, volunteer computing platforms, clouds,… when used for scientific computing. He is one of the main developers of the [**SimGrid**](http://simgrid.gforge.inria.fr/) project, a **simulation** toolkit for building simulators of distributed applications (originally designed for scheduling algorithm evaluation purposes) developed in collaboration with [Henri Casanova](http://navet.ics.hawaii.edu/%7Ecasanova/), [Martin Quinson](http://www.loria.fr/%7Equinson/) and [Frédéric Suter](http://graal.ens-lyon.fr/%7Efsuter/).
###
### Education and Professional Experience
Nov. 2015*Habilitation à Diriger des Recherches*, University of GrenobleOct. 2004-…:Tenured Researcher for the CNRS (*Chargé de Recherche*) at Laboratoire d'Informatique de Grenoble.2004-2005:Post-Doctoral Research Associate, UCSD (USA). Collaboration with Henri Casanova, Larry Carter and Jeanne Ferrante.2003-2004:Post-Doctoral Research Associate, École Normale Supérieure de Lyon (France).2000-Dec. 2003:Ph.D. Computer Science, École Normale Supérieure de Lyon. Laboratoire de l'Informatique du Parallélisme. Thesis: *Heterogeneous parallel algorithms and scheduling : static and dynamic approaches* Advisors: Prof. Olivier Beaumont and Prof. Yves Robert.
Publications
- 59
- 50
- 20
- 9
- 6
- 6
- 6
- 4
- 4
- 3
- 2
- 1
- 1
- 30
- 18
- 8
- 7
- 3
- 5
- 2
- 1
- 1
- 1
- 1
- 1
- 2
- 1
- 6
- 11
- 13
- 14
- 7
- 5
- 2
Centralized Versus Distributed Schedulers Multiple Bag-of-Task ApplicationsInternational Parallel and Distributed Processing Symposium IPDPS'2006, 2006, Rhodes Island, Greece. ⟨10.1109/TPDS.2007.70747⟩
Communication dans un congrès
hal-00789440v1
|
|
Independent and Divisible Tasks Scheduling on Heterogeneous Star-Schaped Platforms with Limited MemoryPDP\'2005, 13th Euromicro Workshop on Parallel, Distributed and Network-Based Processing, 2005, Unknown, pp.179―186
Communication dans un congrès
hal-00789442v1
|
|
Fondements théoriques pour la conception d'algorithmes efficaces de gestion de ressourcesInformatique répartie: architecture, parallélisme et systèmes, 2005, Unknown
Communication dans un congrès
hal-00807012v1
|
|
Pipelining broadcasts on heterogeneous platformsInternational Parallel and Distributed Processing Symposium IPDPS'2004, 2004, Unknown
Communication dans un congrès
hal-00807285v1
|
|
Steady-State Scheduling on Heterogeneous Clusters: Why and How?6th Workshop on Advances in Parallel and Distributed Computational Models APDCM, 2004, Unknown, 171a (8 p.)
Communication dans un congrès
hal-00789446v1
|
|
Optimizing the Steady-State Throughput of Scatter and Reduce Operations on Heterogeneous Platforms6th Workshop on Advances in Parallel and Distributed Computational Models APDCM 2004, 2004, Unknown, 176a (8 p.), ⟨10.1016/j.jpdc.2005.05.021⟩
Communication dans un congrès
hal-00789447v1
|
|
Complexity Results and Heuristics for Pipelined Multicast Operations on Heterogeneous Platforms2004 International Conference on Parallel Processing (ICPP\'2004), 2004, Unknown, pp.267―274, ⟨10.1109/ICPP.2004.1327931⟩
Communication dans un congrès
hal-00789448v1
|
|
Assessing the Impact and Limits of Steady-State Scheduling for Mixed Task and Data Parallelism on Heterogeneous PlatformsHeteroPar\'2004: International Conference on Heterogeneous Computing, Jointly Published with ISPDC\'2004: International Symposium on Parallel and Distributed Computing, 2004, Unknown, pp.296―302, ⟨10.1109/ISPDC.2004.12⟩
Communication dans un congrès
hal-00789444v1
|
|
Scheduling Strategies for Mixed Data and Task Parallelism on Heterogeneous Clusters and GridsPDP'2003, 11th Euromicro Workshop on Parallel, Distributed and Network-Based Processing, Feb 2003, Gênes, Italy. pp.209―216, ⟨10.1109/EMPDP.2003.1183590⟩
Communication dans un congrès
hal-00789453v1
|
|
Optimal Algorithms for Scheduling Divisible Workloads on Heterogeneous SystemsIPDS'2003, Parallel and Distributed Processing Symposium, 2003, Feb 2003, Nice, France. ⟨10.1109/IPDPS.2003.1213202⟩
Communication dans un congrès
hal-00789449v1
|
|
Load-Balancing Iterative Computations on Heterogeneous Clusters with Shared Communication LinksPPAM-2003: Fifth International Conference on Parallel Processing and Applied Mathematics, 2003, Unknown, pp.930―937, ⟨10.1007/978-3-540-24669-5_120⟩
Communication dans un congrès
hal-00789452v1
|
|
Mapping and Load-Balancing Iterative Computations on Heterogeneous ClustersEuro-PVM-MPI-2003: Recent Advances in Parallel Virtual Machine and Message Passing Interface, 2003, Unknown, pp.586―594, ⟨10.1007/978-3-540-39924-7_79⟩
Communication dans un congrès
hal-00789450v1
|
|
Ordonnancement En Régime Permanent Pour Plateformes HétérogènesGRID\'2002, Actes de L\'école Thématique Sur la Globalisation Des Ressources Informatiques Et Des Données, 2002, Unknown, pp.325―334
Communication dans un congrès
hal-00789470v1
|
|
Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor GridsPARA\'02: International Conference on Applied Parallel Computing, Jun 2002, Espoo, Finland. pp.423―432, ⟨10.1007/3-540-48051-X_42⟩
Communication dans un congrès
hal-00789455v1
|
|
A Polynomial-Time Algorithm for Allocating Independent Tasks on Heterogeneous Fork-GraphsISCIS XVII, Seventeenth International Symposium On Computer and Information Sciences, 2002, Unknown, pp.115―119
Communication dans un congrès
hal-00789456v1
|
|
Static Scheduling Strategies for Heterogeneous SystemsISCIS XVII, Seventeenth International Symposium On Computer and Information Sciences, 2002, Unknown, pp.18―22
Communication dans un congrès
hal-00789458v1
|
|
Bandwidth-Centric Allocation of Independent Tasks on Heterogeneous PlatformsInternational Parallel and Distributed Processing Symposium IPDPS'2002, 2002, Unknown
Communication dans un congrès
hal-00789457v1
|
|
Static Scheduling Strategies for Dense Linear Algebra Kernels on Heterogeneous ClustersParallel Matrix Algorithms and Applications, 2002, Unknown
Communication dans un congrès
hal-00789454v1
|
|
Mixed Task and Data ParallelismParallel Matrix Algorithms and Applications, 2002, Unknown
Communication dans un congrès
hal-00789459v1
|
|
Master-Slave Tasking with Heterogeneous Processors2001 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA\'2001), 2001, Unknown, pp.857―863, ⟨10.1109/IPDPS.2003.1213103⟩
Communication dans un congrès
hal-00789462v1
|
|
The Master-Slave Paradigm with Heterogeneous ProcessorsCluster\'2001, 2001, Unknown, pp.419―426, ⟨10.1109/TPDS.2003.1233712⟩
Communication dans un congrès
hal-00789460v1
|
|
Heterogeneous Matrix-Matrix Multiplication, or Partitioning a Square into Rectangles: NP-Completeness and Approximation AlgorithmsEuroMicro Workshop on Parallel and Distributed Computing (EuroMicro\'2001), 2001, Unknown, pp.298―305, ⟨10.1109/EMPDP.2001.905056⟩
Communication dans un congrès
hal-00789461v1
|
|
Heterogeneity Considered Harmful to Algorithm DesignersCluster'2000, 2000, Unknown, pp.403―404
Communication dans un congrès
hal-00789464v1
|
|
Dense Linear Algebra Kernels on Heterogeneous PlatformsParallel Matrix Algorithms and Applications, 2000, Unknown
Communication dans un congrès
hal-00789463v1
|
Parallel AlgorithmsChapman & Hall, pp.337, 2008
Ouvrages
hal-00789466v1
|
|
Algorithmique Parallèle ― Cours Et Exercices CorrigésDunod, pp.288, 2003
Ouvrages
hal-00789467v1
|
Static Data Allocation and Load Balancing Techniques for Heterogeneous SystemsYuen, C. K. Annual Review of Scalable Computing, 4, World Scientific Publishing, pp.1-37, 2002
Chapitre d'ouvrage
hal-00789468v1
|