Skip to main content.

This page’s menu:

Matthieu Roy


Full-time researcher (CNRS).
Dependability Group (TSF).
LAAS-CNRS,
7, av du Colonel Roche,
31077 Toulouse


Contact :
phone (+33) (0)5 61 33 78 13
mail    roy at laas.fr


Research

My research work is focused on distributed algorithms, and consistency problems in distributed or embedded systems.
Currently I'm particularly interested in geo-localised systems, i.e. systems made of mobile entities equipped with positioning hardware, and distributed embedded systems (complex critical systems and Wireless Sensor Networks).

Monitoring of embedded systems

Following previous work on run-time verification of real-time specification (with my previous phD student, Thomas Robert, now with ParisTech), I'm investigating monitoring for distributed and embedded systems, a.k.a. Wireless Sensor Networks. An ANR project on the topics has been submitted (Codename MURPHY), involving LAAS, CNAM, LAMIH, INRIA and SmartGrains.
Current state: a LAAS research report (ref. 10007) is available here

Models and abstractions for geo-aware mobile systems

Finding good abstractions to reason on mobile and geo-localised systems may help to develop applications more naturally, hiding many implementaiton details. A first step in this direction: geo-registers allow to store information cooperatively in an area without infrastructure (the link is a LAAS research report. Parts of this paper have been published in OPODIS 2008). A new and improved version is on tracks.

Mobility-aware testbed

Recent work include the development of a reduced-size test-bed for mobile systems.
Some insights on this emulator (notice that this is not a simulator, it runs real hardware!) can be found on our blog.
Two papers that present the platform will be presented at DSN/WADS 2009 and at ACM PODC 2009 .
Generation of mobility: insights on a program developed for tracking users' activities in so-called ubiquitous systems will be available here soon.

Publications

Distributed Algorithms : Agreement problems & condition-based solutions
ps
pdf 		Achour Mostéfaoui, Sergio Rajsbaum, Michel Raynal, and Matthieu Roy. Condition-Based Consensus Solvability: A Hierarchy of Conditions and Efficient Protocols. Distributed Computing, 2004, vol 17, nr 1, pages 1-20.
pdf

ps 		Achour Mostéfaoui, Sergio Rajsbaum, Michel Raynal, and Matthieu Roy. A Hierarchy of Conditions for Consensus Solvability. In Proceedings of the 20th ACM Symposium on Principles of Distributed Computing (PODC-2001), pages 51-60, Newport, RI, 2001. ACM Press.
bib

www		 Achour Mostéfaoui, Sergio Rajsbaum, Michel Raynal, and Matthieu Roy. Condition-Based Protocols for Set Agreement Problems . In Proceedings of the 16th International Symposium on Distributed Computing (DISC-2002), LNCS, Toulouse, 2002. Springer Verlag.


pdf
 Achour Mostéfaoui, Sergio Rajsbaum, Michel Raynal, and Matthieu Roy. Efficient Condition-Based Consensus. In Proceedings of the 8th International Colloquium on Structural Information and Communication Complexity (SIROCCO-2001), pages 75-291. Carleton University Press, 2001.
bib Achour Mostéfaoui, Sergio Rajsbaum, Michel Raynal, and Matthieu Roy. A Hierarchy of Conditions for Interactive Consistency. In Proceedings of the 7th International Conference on Parallel Computing Technologies (PaCT-2003), Nizhni Novgorod, Russia, September 2003. LNCS.
bib

pdf		 Matthieu Roy and Achour Mostéfaoui. Single Write Safe Consensus Using Constrained Inputs. In Proceedings of the 10th International Colloquium on Structural Information and Communication Complexity (SIROCCO-2003), pages 293-308, Umea, Sweden, 2003. Carleton University Press.

Peer-to-peer computing and dynamic systems
pdf
 L. Courtes, M.O. Killijian, D. Powell, M. Roy : Sauvegarde coopérative entre pairs pour dispositifs mobiles. Ubimob : Deuxièmes Journées Francophones : Mobilité et  Ubiquité, Grenoble, France, juin 2005..
ps Emmanuelle Anceaume, Maria Gradinariu and Matthieu Roy. Self-organizing Systems; Case Study : peer-to-peer systems. Short paper in Proceedings of the 17th International Symposium on Distributed Computing (DISC-2003). Sorrento, Italy.
pdf Emmanuelle Anceaume, Roy Friedman and Maria Gradinariu and Matthieu Roy. An Architecture for Dynamic Scalable Self-Managed Persistent Objects. International Symposium on Distributed Objects and Applications (DOA 2004) – LNCS 3290-3291. CoopIS/DOA/ODBASE (2) 2004: 1445-1462

Distributed Algorithms : Consistency
pdf Michel Raynal, Matthieu Roy, Ciprian Tutu: A Simple Protocol Offering Both Atomic Consistent Read Operations and Sequentially Consistent Read Operations. IEEE Conf. On Advanced Networking and Applications (AINA 2005): IEEE press, pp 961-966. Taïpei, Taiwan.
pdf
 Michel Raynal, Matthieu Roy : Allowing Atomic Objects to Coexist with Sequentially Consistent Objects. Proc. Intl Conf. on Parallel Computing Technologies (PaCT-2005). LNCS 3606, pp 59-73.
bib  www
Divyakant Agrawal, Amr El Abbadi, Achour Mostéfaoui, Michel Raynal, and Matthieu Roy. Towards a Formal Model for View Maintenance in Data Warehouses . In Proceedings of the 21th ACM SIGACT-SIGOPS International Symposium on Principles of Distributed Computing (PODC-2002). ACM Press, Jul 2002. Brief Announcement.
ps 
pdf		 Divyakant Agrawal, Amr El Abbadi, Achour Mostéfaoui, Michel Raynal, and Matthieu Roy. The Lord of the Rings: Efficient Maintenance of Views at Data Warehouses. In Proceedings of the 16th International Symposium on Distributed Computing (DISC-2002), LNCS, Toulouse, 2002. Springer Verlag.

Projects

HIDENETS

HIDENETS is an FP6 STREP project.
I participate in the development of building blocks/abstraction for secure mobile computing.
I am responsible of a demonstrator of the project, which provides a secured, cooperative backup of critical data.

PRIME


PRIME addressed the issues of digital identity management, and privacy protection on internet. I was working on access control mechanisms in privacy-aware databases, and was responsible for a demonstrator, in collaboration with Lufthansa, of an integrated privacy-aware database for Personal Identification Information (PII).

ReSIST NoE


ReSIST (Resilience for Survivability in IST) is a Network of Excellence on the topics of Scalable Resilience. Start date: 2006/01/01.
Within the network, I coordinate a "mini project" that involves IRISA and Univ. Roma, on the topics of abstractions for mobile and localised system.

ASSERT


The ASSERT project was a wide european-funded IP that tackles the issues of critical embedded real-time systems with two objectives: improving the overall chain of development, and providing (some) proof-based software building blocks.

Teaching


ISAE/ENSICA (Toulouse): Operating systems
ENSEIRB (Bordeaux): Dependable Embedded systems
ENSEEIHT (Toulouse): Distributed Algorithms

Links

PhD

    I defended my phD in nov, 2003. The full text is in french only. For the non-french reading persons who might be interested in it, check the above papers on Condition-Based Agreement, or drop me a mail.