» Olivier Perrin

"The whole of science is nothing more than a refinement of everyday thinking" (A. Einstein)

» Research themes

In recent years, I have been particularly interested in everything related to the security of interactions between several services. I have defined an access control model that is compatible with the needs and functioning of service-oriented architectures in a context of cooperative applications. With Ehtesham Zahoor, who became Assistant Professor in Pakistan, we proposed a model that allows us to validate a composition of services against a security policy, and to check whether an execution is correct.

I'm still working on service composition with my work with Anis Ahmed Nacer and François Charoy. We study the catalogue of services provided by service providers (Amazon, Azure, Heroku...) in order to be able to select the best services according to several criteria (performance, security, costs...). The service plans are thus analysed and compared in order to propose the best possible combination according to the targeted architectures and the needs expressed by the architects of the companies using the services.

I addressed with Claudia Ignat and Long Nguyen the issue of trust and privacy in collaborative applications, applications that often go beyond the boundaries of the organization that manages them. We proposed a key transparency system based on the Blockchain that meets the requirements of decentralization, security and ease of use. At the same time, I worked on identifying and detecting attacks on the Blockchain. I am continuing this work in Jean-Philippe Eisenbarth's thesis and in the Fair[&]Smart project that is about to start. Trust is not limited to services, but also to data, for which one must be able to guarantee that they will not be used in an unintended way.

Last, collaborative systems are by nature distributed systems, and I address the issue of data consistency in these systems in Matthieu Nicolas' thesis. To achieve high availability, distributed systems must replicate data and minimize coordination between nodes. CRDTs are data types that support simultaneous modifications and integrate a conflict resolution mechanism. To achieve this, CRDTs attach identifiers to elements stored in the data structure. We propose a new CRDT that integrates a renaming mechanism in order to reduce performance problems (operation, memory) related to the increase in the size of the identifiers.

» PhD Students

My current PhD students are:

My former PhD students are:

» Projects

Here is the list of the projects I was/will be involved in:

» PhD/HDR comittees

I was member of the following comittees:

» List of all my publications

Here is a list of all my publications:

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

· Last update: November, 2019 ·