Français Anglais
Accueil Annuaire Plan du site
Home > Research results > Dissertations & habilitations
Research results
Ph.D de

Ph.D
Group : Parallelism

Stockage fiable des donnés dans les grilles, application au stockage des images de checkpoint.

Starts on 01/10/2006
Advisor : CAPPELLO, Franck

Funding : AM
Affiliation : Université Paris-Sud
Laboratory : LRI

Defended on 16/12/2010, committee :
Rapporteurs:

Gabriel Antaniu, chargé de recherche INRIA.
Pierre Sens, Professeur Université Paris6.

Examinateurs:

Yannis Manoussakis, Professeur Université Paris11 Orsay.
Derrick Kondo, chargé de recherche INRIA.

Directeurs de thèse:

Franck Cappello, Directeur de Recherche Inria.
Thomas Heraults, Maître de conférence, Université Paris-Sud XI

Research activities :

Abstract :
Rollback/recovery solutions rely on checkpoint storage reliability
(after a failure, if the checkpoint images are not available, the
rollback operation fails)
The goal of this thesis is to propose a reliable and an efficient
checkpoint storage service.
By reliable, we mean that whatever the scenario of failures is, as long
as it respects the assumptions made by the algorithms, the checkpoint
images are still available.
And we mean by efficient, minimizing the time required to transfer and
to store the checkpoint images. This will minimize the global execution
time of the checkpoint waves.
To ensure those two points (reliability and efficiency), we propose:
I. A new coordinated checkpoint protocol which tolerates checkpoint
server failures and clusters failures, and ensures a checkpoint storage
reliability in a grid environment.
II. A distributed storage service structured on three layers architecture:\
1. The replication layer: to ensure the checkpoint storage reliability,
we propose to replicate the images over the network. In this direction,
we propose two hierarchical replication strategies adapted to the
considered architecture and that exploit the locality of checkpoint
images in order to minimize inter-cluster communication.
2. The scheduling layer: at this level we work on the storage efficiency
by reducing the data transfer time. We propose an algorithm based on the
uniform random sampling of possible schedules.
3. The scheduling engine: at this layer, we develop a tool that
implements the scheduling plan calculated in the scheduling layer.

Ph.D. dissertations & Faculty habilitations
DECODING THE PLATFORM SOCIETY: ORGANIZATIONS, MARKETS AND NETWORKS IN THE DIGITAL ECONOMY
The original manuscript conceptualizes the recent rise of digital platforms along three main dimensions: their nature of coordination devices fueled by data, the ensuing transformations of labor, and the accompanying promises of societal innovation. The overall ambition is to unpack the coordination role of the platform and where it stands in the horizon of the classical firm – market duality. It is also to precisely understand how it uses data to do so, where it drives labor, and how it accommodates socially innovative projects. I extend this analysis to show continuity between today’s society dominated by platforms and the “organizational society”, claiming that platforms are organized structures that distribute resources, produce asymmetries of wealth and power, and push social innovation to the periphery of the system. I discuss the policy implications of these tendencies and propose avenues for follow-up research.

DISTRIBUTED COMPUTING WITH LIMITED RESOURCES


VALORISATION DES DONNéES POUR LA RECHERCHE D'EMPLO