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

Ph.D
Group : Parallel Systems

Méthodes de génération automatique de code appliquées à l’algèbre linéaire numérique dans le calcul haute performance

Starts on 01/10/2013
Advisor : BABOULIN, Marc

Funding : Contrat doctoral uniquement recherche
Affiliation : Université Paris-Sud
Laboratory : LRI PARSYS

Defended on 26/09/2016, committee :
Directeur de thèse
Marc Baboulin, Professeur, Univ. Paris-Sud, Orsay

Co-encadrant de thèse
Joël Falcou, Maître de Conférences, Univ. Paris-Sud,

Rapporteurs
-Paolo Bientinesi, Professeur, Aachen University, Aachen, Germany
-David Hill, Professeur, Univ. Blaise Pascal, Clermont-Ferrand

Examinateurs
-Frédéric Magoulès, Professeur, Ecole Centrale Paris
-Emmanuel Chailloux, Professeur, Université Pierre et Marie Curie

Research activities :

Abstract :
Parallelism in today’s computer architectures is ubiquitous whether it be in supercomputers, workstations or on portable devices such as smartphones. Exploiting efficiently these systems for a specific application requires a multidisciplinary effort that concerns Domain Specific Languages (DSL), code generation and optimization techniques as well as application-specific numerical algorithms.
In this PhD thesis, we present a method of high level programming that takes into account the features of heterogeneous architectures and the properties of matrices to build a generic dense linear algebra solver. As GPUs have become an asset in high performance computing, incorporating their use in general solvers is an important issue.
We extend our approach to a new multistage programming model that alleviates the interoperability problems between the CPU and GPU programming models.
Our multistage approach is used to automatically generate GPU code for CPU-based
element-wise expressions and parallel skeletons while allowing for type-safe program generation.
Finally, we investigate how to apply high level programming techniques to batched
computations and tensor contractions. We first explain how to design a simple data container
using modern C++-14 programming techniques. Then, we study the issues around batched
computations, memory locality and code vectorization to implement a highly optimized
matrix-matrix product for small sizes using SIMD instructions.

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