Description
Date depot: 1 janvier 1900
Titre: Content aware routing and bandwidth allocation in next generation networks
Directeur de thèse:
Dario ROSSI (LTCI (EDMH))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
Context and State of Art)
Nowadays users have a notion of the Internet as the best way to find information of any kind in the fastest
possible way. Information may be any kind of multimedia content, entertainment, news, technology or business
related information, but is or also a very efficient way to communicate all around the world.
Today Internet is used as a content based network while not being a content centric network. The Internet
architecture has been conceived in order to connect two hosts in the simplest way over any kind of access
technology. The simplicity of providing connectivity allowed Internet to become the largest overlay network ever
deployed. This paradigm never evolved toward a real content centric network while a large set of patches have
been applied to the first original design as ARP, DNS, WEB, CDN, Google in order to make it possible to use the
Internet as it were a real content based network. The number of patches has grown so much that the original
simple design of the Internet has become a very complex system of routers and middleboxes implementing a
large number of functionalities.
In the last years the research literature focused a lot on finding scalable way to tag content with names and
locate it efficiently. An example is given by DONA [1] « Data Oriented Network Architecture» conceived and
developed at the University of Berkeley in California. This type of content naming and location system would
allow to deploy forwarding and routing protocols like those proposed in [2,3].
The paradigm of contentcentric networking has received raising interest especially thanks to the work of Van
Jacobson, the father of Internet congestion control, on «ContentCentric Networking» [4] especially after his
speech at Google Labs in Mountain View in California that has been broadcasted by Google Video [5] and also
a recent interview by Craig Partridge published at ACM Queue.
Today, the challenge to be able to deploy a real content centric network is at the centre of the applied research
on communication network.
Thesis objectifs and technical challanges)
This challenging topic necessitates to be attacked with multiple methodological approaches. The main one would
probably be computer simulation but also, for very simple cases, stochastic optimisation and control.
The candidate should have a solid background in computer science and be comfortable with mathematics, in particular statistics and control theory.
Global planning)
* State of the art (6 months),
* Definition of routing and resource sharing algorithms (2.5 years)
* Performance evaluation and comparison of the chosen algorithms through computer simulation and mathematical modelling, (2.5 years)
* We envisage also to develop a simple proof of concept prototype thank to some internship with student in computer engineering.
Contribution to research projects
Bibliography)
[1] A DataOriented (and Beyond) Network Architecture, T. Koponen, M. Chawla, B. Chun, A. Ermolinskiy, K.
H. Kim, S. Shenker, and I. Stoica Proceedings of ACM SIGCOMM'07, Kyoto, Japan, August, 2007
[2] Forwarding in a ContentBased Network', A. Carzaniga and A.L. Wolf Proceedings of ACM
SIGCOMM'03, Karlsruhe, Germany. August, 2003.
[3] A Routing Scheme for ContentBased Networking, A. Carzaniga, M.J. Rutherford and A.L. Wolf
Proceedings of IEEE INFOCOM'04. Hong Kong, China. March, 2004.
[4] ContentCentric Networking, V. Jacobson, M. Mosko, D. Smetters, J.J. GarciaLunaAceves, Whitepaper
describing future assurable networks, Palo Alto Research Center, January 2007.
[5] Van Jabocson's Google Tech Talk speech entitled « A New Way to Look at Networking» August 30,
2006 http://video.google.com/videoplay?docid=6972678839686672840
[6] A conversation with Van Jacobson and Craig Patridge, ACM Queue, January 2009 http://mags.acm.org/queue/200901/?pg=3D8
Doctorant.e: Gallo Massimo