• Summary
• LIFO Members (Directory)
• Research Teams
• CA
• GAMoC
• LMV
• Pamda
• SDS
• Research Actions
• Doctoral Research
• Seminars and Events
• Seminars (french)
• Events at LIFO
• Events out of LIFO
• Publications and Software
• Publications
• Doctoral and Habil. Theses
• Research Reports
• Software
• Technology transfer

• Contacts/Reaching LIFO
• Who's who at LIFO
• Coming to LIFO
• Teaching
• CS Department
• CS Technical Department
• INSA Centre-Val de Loire

Contact

LIFO - Bâtiment IIIA
Rue Léonard de Vinci
B.P. 6759
F-45067 ORLEANS Cedex 2

Email: contact.lifo
Tel: +33 (0)2 38 41 99 29
Fax: +33 (0)2 38 41 71 37

LIFO working groups (french)

10/04/2018 : [BigData] Convergence HPC et BigData
Sophie Robert (LIFO) Résumé
Débute à 10 h 30.

27/11/2017 : [BigData] Différences de scheduling entre Hadoop V1 et V2.x
Laurent Bobelin (INSA CVL) Résumé

24/05/2017 : [BigData] Tutoriel BigData - Suite et fin
Mostafa BAMHA (LIFO) Résumé
Débute à 10 h. Lieu : Salle de communication

16/05/2017 : [BigData] Tutoriel BigData
Mostafa BAMHA (LIFO) Résumé
Débute à 09 h. Lieu : Espace communication - 1er etage

20/07/2016 : [LMV] Présentation du papier de Aiken & Gray, Barrier Inference, POPL 1998
Wadoud Bousdira (LMV, Université d'Orléans) Résumé
Débute à 14 h 30. Lieu : Salle SR3, bâtiment 3IA

Résumés des groupes de travail

[BigData] Convergence HPC et BigData Sophie Robert, LIFO

[BigData] Différences de scheduling entre Hadoop V1 et V2.x Laurent Bobelin, INSA CVL
Hadoop is one of the major frameworks dedicated to big data distributed computing. Initially dedicated to MapReduce programming framework in version 1, it has turned since version 2 in a generic middleware able to handle any kind of distributed application. In order to do so, Hadoop has known tremendeous changes in its architecture and design since version 1.x. Scheduling is one to the fields of continuous improvement: since the major change of 2.x, the scheduler has changed tremendously both in terms of features, goals and ability to scale. This seminar presents the current Hadoop scheduler architecture and design as well as nowadays mainstream architecture based on separated cluster-wide allocation and per application scheduling.

[BigData] Tutoriel BigData - Suite et fin Mostafa BAMHA, LIFO
Suite du tutoriel sur les outils et frameworks Bigdata pour le traitement de grandes masses de données. Pensez à copier la machine VirtualBox mise à votre disposition sur mirevserveur (pour ceux qui n'etaient present au premier tutoriel).

[BigData] Tutoriel BigData Mostafa BAMHA, LIFO
Tutoriel sur les outils et frameworks Bigdata pour le traitement de grandes masses de données. Pensez à copier la machine VirtualBox mise à votre disposition sur mirevserveur.

[LMV] Présentation du papier de Aiken & Gray, Barrier Inference, POPL 1998 Wadoud Bousdira, LMV, Université d'Orléans
Abstract of the paper: Many parallel programs are written in SPMD style i.e. by running the same sequential program on all processes. SPMD programs include synchronization, but it is easy to write incorrect synchronization patterns. We propose a system that verifies a program's synchronization pattern. We also propose language features to make the synchronization pattern more explicit and easily checked. We have implemented a prototype of our system for Split-C and successfully verified the synchronization structure of realistic programs. DOI: http://dx.doi.org/10.1145/268946.268974