[beasiswa] [INFO] PhD position in France (Data Bases, Security)
From: Francois Lesueur <firstname.lastname@example.org>
Date: Tue, Feb 21, 2012 at 12:31 AM
Subject: [Dbworld] Call for application — PhD position in Lyon, France
Call for application
PhD position in Lyon – France
Keywords : Data Bases, Security
Subject: Access Control for Information Flows in Distributed Environments (see below for more details)
Deadline: March, 2nd 2012
We are seeking for very good candidates in order to apply for a funding by DGA and CNRS-INS2I.
The selection process is two steps. The first step, subject selection, has been passed with success. For the second, and final step, the application file must also present the student who will work on the project. To be clear, the applicant’s quality is a main criteria.
by DGA and CNRS-INS2I
Amount: 1760â‚¬ gross (around 1500 net)/ month for 3 years
The candidate must hold a MSc in computer science, or related field, by september 2012.
Send the following documents (pdf format) to Francois.Lesueur@insa-lyon.fr and Philippe.Lamarre@insa-lyon.fr before March, 2nd 2012.
– Detailed CV including grades and rankings
– motivated application letter.
– support letter
Main selection criteria will be:
– Training quality and results (please provide the ranking…)
– Mastery of english language (written and spoken) ; indicate also the mastery of french language.
Contacts and advisors:
– Philippe Lamarre (Full Professor – LIRIS, INSA Lyon) : Philippe.Lamarre@insa-lyon.fr
– FranÃ§ois Lesueur (Associate Professor – LIRIS, INSA Lyon) : Francois.Lesueur@insa-lyon.fr
Lyon – LIRIS Laboratory (Laboratoire d’InfoRmatique en Image et SystÃ¨mes d’information) – Data Bases team.
– application deadline: March, 2nd 2012
– final notification : June 2012
– Lyon (GoogleMap): http://maps.google.fr/maps?q=lyon
– LIRIS lab: https://liris.cnrs.fr/
* DB Team: https://liris.cnrs.fr/bd/
– Site DGA: http://www.defense.gouv.fr/dga
– Site CNRS-INS2I : http://www.cnrs.fr/ins2i/
These last years have shown the large power owned by a few centralized internet services, such as search engines or social networks. Distributed systems are thus more and more seen as a mean to distribute this power. However, although distributed indexes have been proposed (DHT, Semantic Overlay Network), they are limited to simple requests (for instance, keywords) and complex queries are still reserved to centralized systems. In the QTor project, we aim at extending distributed indexes to complex queries.
The proposed PhD thesis consists in proposing novel mechanisms to control the propagation of the information flows in QTor and, more generally, suited to any distributed environment in which information flows among different authorities. In fact, in a classical centralized system, data publishers enforce themselves access control on their data. In QTor where the queries are evaluated collaboratively, query results can be propagated by intermediate peers ; data publishers must delegate the access control on their data to these intermediate peers. We want to provide this propagation control in a distributed system.
Allowing data publishers to control the propagation of their data involves solving two different problems: the proposition of an access control model taking data propagation into account and the enforcement of this model. We plan to develop both aspects during this thesis.
Data being propagated through different entities, the security model must allow to track information flows: a peer aggregating flows from two different sources must be able to compute the access rights on the result. Previous work proposed models to track flows which are either:
– too constraining to be used in a large scale civilian network (multi-level access control from Biba or Bell and La Padula);
– not expressive enough because based on low-level structures such as files or memory pages (Flume);
– too hard to deploy because needing an ad hoc instrumentation for each application (Fabric).
To the contrary, databases provide a high abstraction level, in particular for access control (access rights on views, role-based models), and seem to be relevant for information flow tracking. The first objective of this thesis is thus to propose a security model suited to track information flows in distributed databases.
Enforcement of security models based on flow tracking in an uncontrolled environment (i.e., a distributed system), which is hard problem, has not been addressed, up to our best knowledge. In fact, once some information has been transferred to another peer, how to keep the control on its diffusion ? This objective may be attained using cryptographic mechanisms (allowing some operations on ciphered data without revealing the clear text), watermarking with a traitor tracing scheme or even trusted code implementing digital restrictions. These methods, used in systems without flow tracking, have not yet been studied with information flows. The second objective of this thesis it thus to propose a solution to enforce the security model in an untrustworthy environment.
Please do not post msgs that are not relevant to the database community at large. Go to www.cs.wisc.edu/dbworld for guidelines and posting forms.
To unsubscribe, go to https://lists.cs.wisc.edu/mailman/listinfo/dbworld
28 total views, 1 views today