The group of Adaptive Distributed Applications & Middleware (ADAM http://adam.lille.inria.fr) at the INRIA Lille - Nord Europe (http://www.inria.fr/lille) is announcing two postdoctoral positions in the area of component models for dependable adaptations:
1- Towards a Formal Theory of Component Models
2- Self-Organizing Large-Scale Adaptations
End of the campaign : June 19, 2009
Beginning of the positions: from September to December 2009
CONTEXT
INRIA, the national institute for research in computer science and control, is dedicated to fundamental and applied research in information and communication science and technology (ICST). Throughout its eight research centres located in seven major regions (Aquitaine, Bretagne, Lorraine, Ile-de-France, Nord Pas-de-Calais, Provence Alpes Cote d'Azur, Rhone-Alpes), the Institute has a workforce of 3,700, 2,900 of whom are scientists from INRIA and its partner organizations. INRIA has an annual budget of 162 million euros, 20% of which comes from its own research contracts and development products. INRIA develops many partnerships with industry and fosters technology transfer and company foundations in the field of ICST - some eighty companies have been funded. Startups are financed in particular by INRIA Transfert, a subsidiary of INRIA that supports four startup funds. The international collaborations are based on an incentive strategy of welcoming and recruiting foreign students as well as developing strong exchanges between research scientists. Priority is given to geographic zones with strong growth: Europe, Asia and North America while maintaining reasonable cooperation with South America, Africa and Middle-East.
ADAM (http://adam.lille.inria.fr) is a project-team of the INRIA Lille - Nord Europe research center (http://www.inria.fr/lille). Members of the ADAM project-team are also part of the LIFL (Laboratoire d'Informatique Fondamentale de Lille) which is a joint unit between CNRS and the University of Lille 1. The objective of the ADAM (Adaptive Distributed Applications and Middleware) project-team is to provide a set of concepts, paradigms, approaches, frameworks, and tools based on advanced software engineering techniques such as CBSE (Component-Based Software Engineering), AOSD (Aspect-Oriented Software Development) or CAC (Context-Aware Computing) to build distributed adaptive software systems (applications and middleware) involving in multi-scale environments and to take into account the adaptation all along the software life-cycle. The ADAM project-team proposes solutions to manage the evolution of application requirements in terms of functional and extra-functional properties either at the level of execution platforms or at the design level. The ADAM project-team applies them to component-based and service-oriented computing distributed applications and platforms.
TOPIC N.1: Towards a Formal Theory of Component Models (contact: Philippe Merle <philippe.merle at inria.fr>)
Component-Based Software Engineering (CBSE) is a world-wide and well-known approach to build configurable and dynamically adaptable software systems. However, there exits a plethora of component models as UML and CCM from OMG, SCA from OASIS, COM/COM+/DCOM/ .NET from Microsoft, JavaBeans/EJB/JMX/JBI from Sun Microsystems, OSGi, the Spring Framework, and a lot of other industrial or academic models. Component models are strongly heterogeneous as each can target a distinct domain of applications and provides specific extra functional properties to applications. For instance, SCA provides a component model for integration in Service-Oriented Architectures (SOA), EJB focuses on secure and transactional enterprise information systems, OpenCom is a generic component model for building systems software, or Fractal is a general-purpose lightweight reflective component model applying to systems-on-chip, operating systems, middleware, application servers, and grid computing. For a same extra functional property, component models can provide different behavioral semantics, e.g., OSGi and Fractal provide different (perhaps incompatible) semantics for the life cycle of components. However, complex software systems are often built as composition of heterogeneous sub-systems, each built on top of one component model. Then, there is a strong requirement to address the composition and interoperability of heterogeneous component models. This encompasses the composition of heterogeneous 1) component-oriented notions (e.g., binding OpenCom receptacles to SCA services) and 2) extra functional behavioral semantics (e.g., unifying the life cycle of OSGi and Fractal).
To address the composition and interoperability of heterogeneous component models, we propose to rigorously 1) formalize component models as mathematical structures, 2) capture common foundations between component models, and 3) compare them in order to identify their equivalences, intersections, complementarities, and differences. At the end, this research must allow us to build a mathematical theory explaining and classifying the notions and semantics of component models formally. The work program for the proposed post-doctoral position is:
1. Selection of component models to formalize. As there is a plethora of component models, the first activity is to select the component models to formalize in next steps. We have a strong interest for the Fractal, OpenCom, SCA, JBI, and OSGi component models but other can be also considered.
2. Formalization of selected component models. This second step consists in the formalization of component models selected in step 1. We strongly encourage using the first-order relational logic and the Alloy formal specification language as formal method and tool respectively.
3. Consolidation of common formal foundations. We recently defined a formal specification of the Fractal component model with Alloy. This specification proposes some common foundations to build component models. This third step consists to consolidate and enhance these common foundations based on feedbacks from work done in step 2.
4. Definition of a formal aspect-oriented architectural definition language. The expected formal ADL should put at work the composition and interoperability between heterogeneous component models by allowing 1) the composition of heterogeneous components in order to select the most appropriate model for each component, and 2) the composition of its own/new component model by weaving concerns coming from different component models.
TOPIC N.2: Self-Organizing Large-Scale Adaptations (contact: Romain Rouvoy <romain.rouvoy at inria.fr>)
Service-Oriented Architectures (SOA) provide a versatile paradigm for building large-scale distributed systems. These systems are generally composed from legacy services published in different domains. However, the variability of the respective hosting infrastructures requires the SOA to be continuously adapted in order to keep satisfying the end-users. As a consequence, the growing complexity of these systems deployed in large-scale environments brings new research challenges in terms of collection, dissemination, and application of distributed adaptations. Indeed, the existing approaches for performing SOA adaptations do not scale and thus only provide limited solutions to the adaptation of large-scale systems. In particular, current approaches to self-adaptation need to be revisited to provide much more flexibility in the way adaptation domains are defined. Adaptation domains control the scope and the visibility of the adaptations that are performed within the domain. However, current approaches define their adaptation domain based on static criteria (e.g., network partition, application partition), thus leading inevitably to inefficient adaptations on the long term.
Therefore, the objective of this postdoc is to propose and implement a middleware solution supporting the organization, collection,dissemination, and application of large-scale adaptations. This includes i) the definition of adaptation domains, ii) the deployment of context information collectors, iii) the dissemination of contextual situations collected in the domain, and iv) the dissemination of adaptation decisions within the adaptation domain. To achieve this objective, we are particularly interested in combining the principles of fuzzy logic and group communication systems. In particular, we believe that the combination of these approaches can provide a flexible solution to the definition of adaptation domains by using the concept of fuzzy group, which can dynamically shrink, expand, split, or merge depending on ad hoc criteria.
The work to realize during this thesis will be organized as follows:
- Studying the state-of-the-art in the domain of large-scale adaptations. The study will particularly focus on the organization and scalability of the related works in this domain;
- Propose a comprehensive model for the self-organization of distributed adaptations. In particular, it will identify the key criteria for organizing the self-adaptation of large-scale systems. This model will also exhibit autonomous capabilities and will not rely on any centralized infrastructure or knowledge;
- Implement a middleware solution enabling the self-organization of distributed adaptations based on the proposed model. This distributed middleware will dynamically control the partition of adaptation domains and assign responsibility to nodes within each adaptation domain;
- Validate the proposed solution on large-scale adaptation scenario. The target infrastructure for demonstrating the results will be a grid or ubiquitous environment.
More information at: http://www.lifl.fr/~rouvoy/topics/postdoc-cappucino-09.html
APPLICATION REQUIREMENTS
Applicant should have held a doctorate or Ph.D. for less than one year or you are about to obtain one and you would like to carry out a fulfilling research activity in the field of ICST (information and
communication science and technology) or in a related field.
POSITION SALARY
2,357.30 EUR gross per month
SOCIAL SECURITY BENEFITS
- Entitled to unemployment benefit at the end of the contract;
- Affiliated to the French social security system.
HOW TO APPLY
Thank you for applying directly on the institute's website, by following this link: http://www.inria.fr/travailler/o ... doc/postdoc.en.html |
登录
