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

Ph.D
Group : Parallel Architecture

Emugrid: study of virtualization mechanisms for large scale Grid emulation

Starts on 01/09/2004
Advisor : CAPPELLO, Franck

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

Defended on 12/09/2008, committee :
Rapporteurs
PLATEAU Brigitte
SENS Pierre

BEAUQUIER Joffroy
CAPPELO Franck
LEFEVRE Laurent
MANOUSSAKIS Yannis
RICHARD Olivier

Research activities :

Abstract :
In distributed system's domain, emulation is a significant methodology. It allows to test real applications, observe their behavior, to experimentally verify some of their properties and to achieve measurements in a controlled environment. However, there is no emulator for the large scale which owns essential characteristics like reproducibility.

The main goal of this thesis is to make up an environment to emulate large scale distributed systems (typically 100 000 nodes). The accomplished work has been first to enumerate the different ways for evaluating distributed system at large scale. Then, we have defined expected characteristics for a tools which could emulate many machines while abide with essentials characteristics like fairness, reproducibility, etc. This specification is a at the root of a architecture proposal for the V-DS tool (Virtualization environment for large-scale Distributed Systems).

The continuation of this thesis describes implementation choices et the software / middleware selection to realize the emulator. A set of experiments achieves with 10 000 virtual machines on a 250 nodes cluster with bi-processors validates initially defined properties.

The validated emulator has been used to study virtualization impact on parallel application execution. Experiments have mainly consist in measuring execution time of MPI (Message Passing Interface) processes by varying the amount of virtual machines by physical machines. V-DS has also permit us to test how scale different DHT (Distributed Hash Tables) implementations. At last, the emulator is actually been upgraded to allow “data link” network layer emulation.

Ph.D. dissertations & Faculty habilitations
CAUSAL LEARNING FOR DIAGNOSTIC SUPPORT


CAUSAL UNCERTAINTY QUANTIFICATION UNDER PARTIAL KNOWLEDGE AND LOW DATA REGIMES


MICRO VISUALIZATIONS: DESIGN AND ANALYSIS OF VISUALIZATIONS FOR SMALL DISPLAY SPACES
The topic of this habilitation is the study of very small data visualizations, micro visualizations, in display contexts that can only dedicate minimal rendering space for data representations. For several years, together with my collaborators, I have been studying human perception, interaction, and analysis with micro visualizations in multiple contexts. In this document I bring together three of my research streams related to micro visualizations: data glyphs, where my joint research focused on studying the perception of small-multiple micro visualizations, word-scale visualizations, where my joint research focused on small visualizations embedded in text-documents, and small mobile data visualizations for smartwatches or fitness trackers. I consider these types of small visualizations together under the umbrella term ``micro visualizations.'' Micro visualizations are useful in multiple visualization contexts and I have been working towards a better understanding of the complexities involved in designing and using micro visualizations. Here, I define the term micro visualization, summarize my own and other past research and design guidelines and outline several design spaces for different types of micro visualizations based on some of the work I was involved in since my PhD.