[ecoop-info] PhD proposal - INRIA Lille
Anne-Francoise Le Meur
Anne-Francoise.Le-Meur at lifl.fr
Tue Feb 23 16:13:12 CET 2010
[ Apologies for multiple copies of this message ]
The ADAM team-project from the INRIA research institute at Lille in
France is looking for a PhD candidate in the domains of software
architecture and component-based applications. Details about the topic
can be found below.
Applications must be submitted online, the deadline is *May 4*, 2010.
*Title : *An agile framework to support the safe runtime evolution of
Many applications must evolve during their whole life-cycle to cope with
new user requirements, execution context changes, or maintenance tasks.
Unfortunately, any modification at execution time is likely to
jeopardize the robustness of the whole application. It is thus necessary
to provide some guaranties that a given modification will not break the
application properties, preventing then the application from a loss of
quality of service, or even, in the worst case, a system failure.
To reduce software development costs and enhance the flexibility and
maintainability of systems, component-based software engineering (CBSE)
proposes to build software systems by assembling software components. In
this approach the overall structure of the application is first
described with an architecture description language (ADL) . Such
description not only highlights the needed components and their assembly
but also facilitates the understanding and analysis of the application's
properties, such as behavioral or quality of service properties.
Verifying that architectural specifications are coherent at design time
is nevertheless not sufficient to guarantee the robustness of the
application during an evolution at runtime.
The CALICO has been developped by the ADAM INRIA team-project to propose
an approach to incrementally and iteratively design robust
component-based applications through the use of models and the
verification of these models, which combines static analysis of the
application specifications but also dynamic testing at runtime [2, 3, 4,
5]. The approach relies on a strong coupling of the architecture models
and the running application. Nevertheless keeping an abstract
representation of an application at runtime in order to support safe
runtime evolution brings many challenges that need to be addressed.
The goal of this PhD is to develop an approach where architecture models
are used through the whole application life-cycle, enabling safe
evolution at execution time. Any evolution, may that be an adaptation
due to execution context changes or a maintenance task, must take into
consideration all or parts of the application's properties in order to
preserve the application robustness. The PhD research work will identify
which information should be kept in the architecture models, as well as
how this information should be organized in order to efficiently support
the verification at runtime of the properties guaranteeing the
application robustness. Furthermore, since some of this information will
need to reflect the current state of the running application, it will be
thus necessary to provide non intrusive monitoring support and find good
tradeoff to address the communication cost that the monitoring [6, 7]
and information reification induce.
The expected results of this PhD work are the definition of an
architecture-based monitoring and safe-evolution model, and the
development of an infrastructure that will complement the current CALICO
approach to form an agile framework for the safe design and evolution of
component-based applications during their whole life-cycle.
The PhD candidate should:
- Be knowledgeable in the domains of Software Engineering, models, CBSE,
ADL, middleware, distributed systems.
- Have very good programming skills in JAVA
 A classification and comparison framework for software architecture
description languages. N. Medvidovic and R. N. Taylor. In IEEE
Transactions on Software Engineering, volume 26, page 23, jan 2000
 Architectural Specification and Static Analyses of Contractual
Application Properties. G. Waignier, A.-F. Le Meur and L. Duchien. In
Proceedings of the 4th International Conference on the Quality of
Software-Architectures (QoSA 2008), Karlsruhe (TH), pages 152-170,
Germany, oct 2008
 A Model-Based Framework for Statically and Dynamically Checking
Component Interactions. G. Waignier, P. Sriplakich, A.-F. Le Meur and L.
Duchien. In Proceedings of the ACM/IEEE 11th International Conference on
Model-Driven Engineering Languages and Systems (MODELS 2008), pages
371-385, Toulouse, France, oct 2008
 A Model-Based Framework to Design and Debug Safe Component-Based
Autonomic Systems. G. Waignier, A.-F. Le Meur and L. Duchien. In
Proceedings of the 5th International Conference on the Quality of
Software-Architectures (QoSA 2009), pages 1-17, Pennsylvania, USA, jun 2009
 WildCAT: a generic framework for context-aware applications. P.-C.
David, T. Ledoux. In the 3rd International Workshop on Middleware for
Pervasive and Ad-Hoc Computing (MPAC 2005), ACM Digital Library,
Grenoble, France, November 2005.
 Scalable Processing of Context Information with COSMOS. D. Conan,
R. Rouvoy and L. Seinturier. In 7th IFIP International Conference on
Distributed Applications and Interoperable Systems (DAIS'07). p.
210--224 of LNCS 4531 (Springer). Paphos, Cyprus. June 5--8, 2007.
*How to apply: *
Candidates should provide a CV, a motivation letter highlighting their
research statement and interest in this specific PhD topic, and at least
two references. This PhD thesis is proposed in the context of an INRIA
fellowship. Only applications submitted on the INRIA website will be
considered -- see the link "apply online" at the bottom left of this
For more information please contact Anne-Françoise Le Meur
<mailto:Anne-Francoise.Le_Meur at inria.fr>.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the ecoop-info