Direction des Relations Européennes et Internationales (DREI)

Programme INRIA "Equipes Associées"

I. DEFINITION

EQUIPE ASSOCIEE	ASA (Adaptive SoftwAre)
sélection	2007

Projet INRIA : ARLES	Organisme étranger partenaire : Dip. Informatica-Università dell’Aquila
Unité de recherche INRIA : Rocquencourt Thème INRIA : Com A	Pays : Italie

	Coordinateur français	Coordinateur étranger
Nom, prénom	Issarny Valérie	Inverardi Paola
Grade/statut	DR2	Professor
Organisme d'appartenance (précisez le département et/ou le laboratoire)	INRIA	Dipartimento di Informatica, Università dell’Aquila
Adresse postale	INRIA-Rocquencourt Domaine de Voluceau, BP105 78153 Le Chesnay Cedex	Via Vetoio, 1 Coppito L’Aquila
URL	· http://www-rocq.inria.fr/arles/ · http://www-rocq.inria.fr/arles/members/issarny.html	· http://www.di.univaq.it · http://www.di.univaq.it/inverard/paola.html
Téléphone	+33139635717	+390862433127
Télécopie	-	+390862433131
Courriel	Valerie.Issarny@inria.fr	inverardi@di.univaq.it

La proposition en bref

Titre de la thématique de collaboration:

· Assisting the development of dynamic distributed software systems for next generation ubiquitous communication and computing infrastructures

· Support au développement de systèmes logiciels distribués dynamiques pour les futurs environnements de communication et de calcul, ubiquitaires

Abstract: Software in the near ubiquitous future (Softure) will need to cope with variability, as software systems get deployed on an increasingly large diversity of computing platforms and should further deliver applications ubiquitously. Heterogeneity of the underlying communication and computing infrastructures, mobility and continuously evolving requirements demand new software paradigms that span the entire life-cycle, from development to deployment and execution. Softure must be developed in a way that facilitates both its deployment over heterogeneous networks of heterogeneous nodes, and its interaction with end users, their environment and/or other existing systems, depending on the application domain. Moreover, Softure should be reliable and meet the user’s performance requirements and needs. Last but not least, Softure should be dynamic so that the applications they implement can be provisioned ubiquitously, despite the high dynamics of the pervasive networking and computing environment. Looking at the software life cycle, one key issue in this domain appears to be the disappearance of a clear distinction between static and dynamic aspects. Indeed, the adaptability requirement imposed by ubiquity makes software become “evolving” in nature, therefore introducing a strong interaction between the development environment and the middleware one. Goal of the proposed ASA team is to research design and programming techniques and innovative middleware models that can be profitably integrated to support this new generation of software systems.

Présentation de l'Équipe Associée

(environ 2 pages)

1. Présentation du coordinateur étranger

Paola Inverardi is full professor in Computer Science at University of L'Aquila, Italy, since 1994. Previously she has worked at IEI-CNR in Pisa (1984-1994) and at Olivetti Research Lab. in Pisa, Italy (1981-1984). She has been coordinator of the Laurea Program in Computer Science from 1994 to 2000. Since its creation, in 2001, she is Head of the Department of Computer Science at University of L'Aquila, where she leads the Software Engineering and Architecture (SEA) Research Group. She is member of the Senate Board of University of L'Aquila. Her main research area is in the application of formal methods to software development. Her research interests primarily concentrate in the field of software architectures. She has actively worked on the verification and analysis of software architecture properties, both behavioral and quantitative for component-based, distributed and mobile systems. She has served as general chair, program chair and program committee member for many international conferences. She is currently Chair of the Steering Committee of the European Software Engineering Conferences (ESEC) and Chair of the Steering Committee of the Workshop on Software and Performance. Since July 2005 she is member at large of the ACM SIFSOFT (ACM Special Interest Group on Software Engineering) Executive Committee (http://www.sigsoft.org/about/execComm.htm). She is associate editor of the ACM TOSEM Journal. She is leading collaborative projects with several industries in the telecommunication field, among which Siemens, ERICSSON, MARCONI, on topics related to the study of software architectures. She has had and has several collaborations with academic and industrial institutions all over the world, among which University of California, Irvine, University of Colorado, Boulder, Carnegie Mellon University, University of Oregon, Eugene, AT&T Bell Lab., University of Buenos Aires. She has (co-)authored over one hundred publications in international journals and international conference and workshop proceedings.

2. Historique de la collaboration

2.1. entre les équipes

The Software Engineering and Architecture (SEA) Research Group at the University of L’Aquila and the ARLES group at INRIA-Rocquencourt have been interacting for several years due to common research interests in the areas of software architectures and distributed software systems. Close collaboration between the two groups officially started in 2004 with the 4 month visit of Prof. Paola Inverardi at INRIA-Rocquencourt from September to December 2004. The collaboration is since ongoing with joint research work on dynamic software architectures for ambient intelligence/pervasive computing. This effort has in particular led us to initiate research on the theme of adaptive software systems, leading to the proposed ASA team on “Adaptive Software”. As part of this joint effort, Mauro Caporuscio, who got his PhD on event-based systems at the University of L’Aquila under the supervision of Prof. Paola Inverardi, is carrying post-doctoral research within the ARLES team since January 2006. Mauro is more specifically studying development support for dynamic software system architectures for beyond 3G (B3G) networks. Collaboration between the two groups further led to define a European research project on the study of development support (addressing system design, runtime and validation) for adaptive systems in pervasive computing environments enabled by B3G networks. The project, called PLASTIC, is funded by the EC as a STREP IST project since February 2006 for a duration of 30 months (see http://www.ist-plastic.org/); the project is coordinated by Valerie Issarny, while Paola Inverardi acts as coordinator proxy. Joint research work has so far led to the following joint publications:

· Nikolaos Georgantas, Paola Inverardi, Valérie Issarny. Software Platforms. In E. Aarts, J. Encarnacao (editors), True Visions: Tales on the Realization of Ambient Intelligence. Springer Verlag. 2006.

· Antonia Bertolino, Wolfgang Emmerich, Paola Inverardi, Valerie Issarny. Softure: Adaptable, Reliable and Performing Software for the Future. In Proceedings of the 1st International EASST-EU Workshop on Future Research Challenges for Software and Services (FRCCS'06). Associated to ETAPS'06. April 2006, Vienna, Austria.

· Marco Autili, Sonia Ben Mokhtar, Mauro Caporuscio, Nikolaos Georgantas, Paola Inverardi, Valerie Issarny. A Reference Model for Context-aware Adaptable Applications in B3G Networks. Submitted to IEEE/ACM International Conference on Software Engineering (ICSE).

2.2. entre l'INRIA et l'organisme partenaire

In addition to the above collaboration between the SEA and ARLES research groups, the Department of Computer Science of the University of L'Aquila collaborates since 1995 with INRIA in different research areas:

· Formalization and verification of complex systems based on theorem proving techniques: The researcher involved on the INRIA side is Laurent Thery, who has regularly visited the University of L'Aquila since 1995. Starting from 2001, he has also been teaching in the Laurea in Computer Science program. From December 2002 till November 2004, he was "en détachement" from INRIA at the University of L'Aquila, funded by the Italian Ministry of Research.

· Networks communication: The group involved at INRIA is MASCOTTE, led by Jean-Claude Bermond. Since 1996, researchers from the University of L'Aquila, Prof. Flammini and more recently Dott. Navarra, have been regularly visiting INRIA under different funding sources.

· Component based systems: Massimo Tivoli has spent 1 year post-doc in Grenoble, ended October 2006, within the INRIA group POP ART of INRIA-Rhone Alpes.

3. Impact

3.1. sur la collaboration déjà existante avec votre partenaire

The ASA joint research team will significantly strengthen the existing collaboration between the two groups, in particular facilitating visits of researchers between the two groups, and, further, exchanges with researchers of other groups, and possibly other institutions, working in the specific area. This will in particular lead to timely collaborative research work, and hence to timely results in the area of adaptive software systems development, as well as effective dissemination of the work.

3.2. sur la collaboration avec d'autres projets INRIA

The development of dynamic adaptive systems is a key research topic for next generation software systems that will enact the vision of ambient intelligence and/or pervasive computing. This, in particular, involves making the software systems as transparent as possible to end-users despite the high dynamics of the networked resources, thus raising the issue of software system robustness. Hence, the research undertaken in ASA, which focuses on the engineering of adaptive software systems, can lead to fruitful collaboration with other INRIA teams such as teams investigating software system dependability and/or software systems for next generation networking infrastructures.

3.3. sur la collaboration avec d'autres équipes de l'organisme étranger partenaire.

In the same way as for potential collaboration with other INRIA teams, collaboration with other research groups of the Department of Computer Science of the University of L’Aquila may indeed be favored by the ASA research, due to the various research topics that relate to both our specific research and research of other groups for which collaborative research may be beneficial for all parties.

4. Divers

It is important to note that while the IST PLASTIC project conveniently supports collaboration between the SEA and ARLES groups, the research work that we plan to investigate within ASA is longer term and goes beyond topics investigated within PLASTIC. Specifically, PLASTIC is focused on service technology for B3G networks and as such, will provide adequate concrete case studies for ASA research

II. PREVISIONS 2007

Programme de travail

· Survey of adaptive systems and elicitation of appropriate paradigms: A first activity will be devoted to survey the literature on adaptive systems that span different research domains from operating systems to biologically-inspired computational systems. The survey will analyze systems from the software architecture perspective. We recall here that a software architecture is defined as the structure of the components of a system, their interrelationships, principles and guidelines governing their design and evolution over time, along with a set of connectors that mediate communication, coordination or cooperation among components. Thus, software architectures seem to be the right system abstraction to model and reason about evolutionary, dynamic adaptive systems. Software architectures for self-adaptive systems are emerging in the literature. However, proposed approaches still lack associated design and validation methodologies and further address very specific adaptation with respect to computer-centric context awareness. Comprehensive design and validation of self-adaptive systems requires accounting for both functional and non-functional properties and for the highly dynamic pervasive computing and networking environment. In addition, adaptation applies to both the application and middleware layers, regarding both the overall distributed systems and embedded component systems. Thus the study of adaptive systems should account for modelling both the system software architecture and the middleware, trying to reconcile the static, development-time view of a system with the dynamic, run-time view.

· Elicitation of relevant self-properties: The above survey activity will allow the elicitation of a set of requirements that qualify and quantify the dynamic behavior of an adaptive system and its ability to dynamically adapt its functionalities depending on the changes in the executing context and on the variable resource needs. The ability to perform adaptation plays a primary role in providing an appropriate and stable output, and in guaranteeing correct behaviors with respect to functional (e.g., interaction patterns) and non-functional properties (e.g., security).

· Case study in the concrete domain of adaptive service technology for pervasive computing: Pervasive networking provides mobile users with unique features for seamlessly accessing networked services. However, service provisioning faces numerous challenges, among which are: developing services that can be easily deployed on a wide range of evolving infrastructures, from networks of devices to standalone wireless resource-constrained handheld devices; making services resource-aware so that they can benefit from networked resources and related services; and ensuring that users always experience the best Quality of Service (QoS) possible according to their specific situation. Our approach will be based on service technologies, making services adaptable with respect to the networked services with which they interact. This approach shall allow the development of services for diverse application domains, exploiting the significant number and diversity of wireless resources that are (expected to be) networked, thanks to next generation, B3G networks.

· 2008 and beyond: Building upon the above joint research work to be carried out in 2007, our work in 2008 and beyond will relate to investigating further methods, tools and middleware supporting the development of adaptive software systems, focusing on the self-properties to be maintained by the adaptive software, which we will have elicited in the first year.

Budget prévisionnel 2007

1. Co-financement

The ASA team will indirectly benefit from the EC funding of the FP6 IST STREP PLASTIC project (see above). However, as already stated, PLASTIC will not fund ASA research, which is much longer term.

Funding from the University of L’Aquila to support ASA will be mostly indirect, i.e., through support of short-term visits of researchers from University of L’Aquila, as has already been the case in the last couple of years.

2. Echanges

ESTIMATION DES DÉPENSES		Montant
	Nombre	Accueil	Missions	Total
Chercheurs confirmés	4	· INRIA · L’Aquila	· Survey and assessment of existing approaches to support development of adaptive software systems · Study on the modeling of adaptive systems	· 2 short-term visits of about 1 week of INRIA permanent researchers at L’Aquila = 3000€ · 2 short-term visits of about 1 week of L’Aquila permanent researchers at INRIA = 3000€
Post-doctorants	2	· INRIA · L’Aquila	· Assessing distributed middleware architectures with respect to enabling system adaptation · Study on the modeling and monitoring of the QoS-related behavior of systems for guiding self-adaptation	· 2 mid-to-short-term visits of INRIA post-doc researchers at L’Aquila = 5000€ · 2 mid-to-short-term visits of L’Aquila post-doc researchers at INRIA = 5000€
Doctorants	2	· INRIA · L’Aquila	· Study on the modeling and reasoning about adaptive systems · Study of middleware support for adaptive systems	· 2 mid-to-short-term visits of INRIA PhD students at L’Aquila = 5000€ · 2 mid-to-short-term visits of L’Aquila PhD students at INRIA = 5000€
Stagiaires	2	· INRIA · L’Aquila	· Study on the modeling of software architectures of adaptive systems · State of the art survey and analysis of adaptive middleware architectures	· Student of Master 2 Research at University of L’Aquila – Stage of 6 months (15000€) · Student of Master 2 Research at INRIA - Stage of 6 months (15000€)
Autre (précisez) :			· Organization of an international closed workshop by invitation on the engineering of adaptive software systems	· Support to workshop organization and to participation of ASA researchers (6000€)
Total
		- Total des co-financements		· 47000€
	Financement « Équipe Associée » demandé			· 20000€

Remarques ou observations : In the above table, we detail the topics that will be investigated as part of the “missions” that are already foreseen in the context of our research work in 2007. However, further topics/missions might emerge, based on the outcome of our research. Also, in addition to supporting joint research work between the 2 groups of ASA through funding of short-to-mid-term visits, ASA will also support exchanges between researchers working in the area of adaptive systems through the organization of the above mentioned workshop dedicated to the topic and bringing together researchers from the University of L’Aquila, INRIA but also other institutions.