Ce parcours a été remplacé par le parcours UNC en 2009-10
Previous description :
Responsable : Michel Riveill
The aim of this program is to provide excellent academic or industrial career opportunities by offering high level coverage of networking principles that will help to meet the challenges and make the technological choices of tomorrow in the domains of the Internet, Telecommunications and the Distributed Systems. This program proposes courses providing basic and specialisation skills in the area of Network Architectures and Protocols and their evolution, Networks modelling and performance evaluation, Networking and telecommunications algorithms, Peer to peer networking, and Security.
This program are also to train qualified specialists to respond to the strong demand of industry for software engineers, information system designers, and technical analysts and advisers. Also to prepare candidates for further investigation and research into Ambiant, Grid and Network Computing.
Ambient computing: The main goal of ambient computing sometimes called "ambient intelligence" is the creation of intelligent environments, by an implicit assistance of everybody in its day life. Ordinary object facilities are increased using invisible computers embedded in everyday objects and communications between them. The first goal of this set of lectures is to present the fundamental principles behind ambient computing and various middlware approaches to create innovative intelligent services and applications. These principles will be illustrated using very concrete examples. Implementation details will be considered. The second goal of this set of lectures is to detail some solutions to support communication facilities between communicating objects used to implement ambient computing applications. In this area we focus mainly on the notion of ad’hoc network, which are based on short distance wireless communication technologies. The structure of these networks is fully dynamic, depending on the mobility of nodes. We detail some routing protocols, which take into account such as dynamicity. We will consider also the problem of the energy consumption behind the implementation of ad hoc protocols.
Grid computing: Applications are more and more demanding in terms of computing power and storage capacity. Aggregating computational and storage resources is a way to meet the requirements of these highly demanding applications. A grid can be defined as set of resources (processors, memories, disks, networks, instruments, I/O devices...) distributed in several sites, administrative domains or institutions and shared to be used in a coordinated way. Grids are characterized by their large scale in terms of number of resources, number of users and geographical distribution and by the heterogeneity and dynamicity of their resources. The objective of this teaching unit is to present systems, middleware and programming models easing Grid use, management and programming while ensuring efficient, secure and reliable application execution. We focus more specifically on computing grids federating several clusters located in different sites. This leads us to also study cluster system services. Lectures are illustrated by a number of case studies.
Network computing: The future of the Internet is the object of extraordinary academic and industrial investments worldwide. The goal of research activities in this area is to design new algorithmic and architectural bases for networking services and protocols that will allow coping with the weaknesses of the current Internet in security, quality of service, mobility and interconnection with the real world (Internet of the Things). Future networks will most probably be made of very large scale, flat structures: mobile ad hoc networks, sensor networks, Wifi meshes, peer-to-peer applications to quote a few.