Jérome Francois

Summary
Student:	Jérome Francois
Title:	Malware Inspired Network Management
e-mail:	jerome.francois@loria.fr
Affiliation:	INRIA
Supervisor:	Olivier Festor and Radu State
Committee:
Start:	2006
End:	2009
Funding:	CNRS - Lorraine Region

Biography

Due to the rapid evolution of networks and above all the Internet, a huge number of various network equipments need to be managed to assure the well functioning of the network. It becomes a real challenge as the management domain boundaries are not well delimited. A company has to be able to manage all the devices even if they are located on several sites. Thus the management operations has to be transmitted through the Internet and to pass through many active equipments like firewalls or network addresses translators. Many solutions have been proposed to deal with a scalable network management but they are limited by the fact that they need dedicated equipments and many resources. However, the attackers faced with the same problem and they overcame them by using innovative techniques. For example, a worm can infect many computers in few time in order to install a software on each of them. Then, the attacker is able to control all of them in an efficient way to perform distributed denial of service attacks or retrieve sensitive data. This threat is well known and this kind of compromised computers are named botnets. Therefore the worms and the botnets are good examples of scalable and efficient communication mechanisms. The key idea of this thesis is to evaluate if adapting malware techniques is a potential solution for a large scale management. Because only attacks experience are known, using malware techniques needs to be motivated by a proof of their efficiency. Therefore the main objective of the thesis is to propose malware based solutions for network management, model them to evaluate them and determine if they are feasible. The next step is the implementation of these solutions in a real environment with a real management application. Moreover, by studying the malware mechanisms, the second objective is to improve the malware detection techniques. Particularly, the botnets are a major Internet threat and pose serious problems. Therefore, botnet tracking is a real challenge to reduce the impact of them and the associated financial loss. Thus, this thesis will use simulation techniques and real networks for experiments like PlanetLab.

PhD project description

References

1. J. François, R. State, and O. Festor, “Botnet based scalable network management,” in Managing Virtualization of Networks and Services, 18th IFIP/IEEE Distributed Systems: Operations and Management (DSOM 2007), Lecture Notes in Computer Science (LNCS), Vol. 4785, Silicon Valley, CA, USA, October 2007. (pp1-12)
2. J. François, R. State and O. Festor, “Malware Models for Network and Service Management,” in Inter-Domain Management, First International Conference on Autonomous Infrastructure, Management and Security (AIMS 2007), Lecture Notes in Computer Science (LNCS), Vol. 4543, Oslo, Norvège, Juin 2007. (pp192-195)

External links

• [_URL_ Homepage] of Jérome Francois
• Publications of Jérome Francois, as indexed by DBLP