Contact
 
Workshop collocated with
the 19th European Conference on Software Architecture ECSA 2025



Keynote presentation by Professor Nabil Abdennadher, University of Applied Sciences and Arts, Western Switzerland (HES-SO)


Personal Page

Biography:

Nabil Abdennadher received the Diploma in Computer Engineering from Ecole Nationale des Sciences de l’Informatique (ENSI, Tunisia), and the Ph.D. degrees in Computer Science from Univ. of Valenciennes (France) in 1988 and 1991, respectively. He was assistant professor at the Univ. of Tunis II from 1992 to 1998 and research assistant at the Swiss Federal Institute of Technology (EPFL) from 1999 to 2000. In 2001, he joined the University of Applied Sciences, Western Switzerland (HES-SO, HEPIA) as assistant professor. In 2008, he became associate professor and in 2017 he was promoted to full professor. Nabil Abdennadher was head of the inIT research institute at HES-SO, HEPIA from 2010 to 2022. He is currently head of the Large-Scale Distributed Systems research group, representative of the Innovation DataBooster initiative in Swiss Romandie, member of the Editorial Board of the Journal of Reliable Intelligent Environments and member of the Swiss AI centre team. He is author and co-author of several publications and book-chapters. He is currently working on several Swiss and European projects aiming at developing AI-based tools for energy sector.

Title: Towards a self-adaptive Federated Learning (FL) based edge-to-edge framework. Application to energy sector

Abstarct: New forms of electricity production and consumption are disrupting the energy market. Renewable energy sources are intermittent and distributed in vast numbers of small-scale decentralized energy producers, which affects grid management. In parallel, consumption scenarios are changing as fossil fuels are being replaced by electricity in several sectors such as electric vehicles. Along the same lines, digital transformation is reshaping the electricity market. To name a few, internet of things, edge-to-cloud computing and machine learning techniques will allow the proliferation of intelligent versions of today's “smart meter”. We can expect that, soon, the new smart meters devices will support additional functions including:

  • Monitoring: monitor/control home appliances, optimise local consumption or microgrid’s consumption,
  • Learning and predicting consumption/production,
  • Gathering information related to other households
Thanks to these new functionalities, we can expect new “digital energy services” to emerge. These services will rely on a self-adaptive FL-based edge-to-edge framework. This framework must:
  • support IT technology such as Internet of Things (to collect and gather data), Artificial Intelligence (to set up and self-adapt predictive models), Cloud and Edge computing (to store and process data), collaborative models to insure collaboration, etc.
  • unify and optimise computation and data across diverse and distributed resources from From IoT devices to edge devices to Cloud and HPC. This is what is known today as Continuum Computing.
This talk will present the architecture of a self-adaptive FL-based edge-to-edge framework applied to energy sector.

Introduction

The goal of this workshop is to bring together researchers and practitioners both from the Academia and from the Industry working in the areas of the adaptation and reconfiguration of distributed systems. Different investigation topics are involved, such as: Component Based Software Engineering, Web service, cloud applications, mobile applications, Functional and non-functional requirements (QoS, performance, resilience), monitoring, diagnosis, decision and execution of adaptation and reconfiguration. Different research areas are covered: concepts, methods, techniques, and tools to design, develop, deploy and manage adaptive and reconfigurable software systems.

The concept of adaptive and reconfigurable software systems has been introduced in order to describe architectures, which exhibit such properties. An adaptive and reconfigurable software system can repair itself if any execution problems occur, in order to successfully complete its own execution, while respecting functional and non-functional agreements. In the design of an adaptive and reconfigurable software system, several aspects have to be considered. For instance, the system should be able to predict or to detect degradations and failures as soon as possible and to enact suitable recovery actions.

Topics
For this workshop, contributions are devoted to functional and non-functional adaptability and reconfiguration management in service-oriented and component-based software systems. Specifically, the relevant topics include, but are not limited to:
  • Distributed and centralized collaborative solutions for the diagnosis and repair of software systems
  • Design for the diagnosability and repairability
  • Collaborative Management of Non-Functional requirements (quality, security, robustness, availability)
  • Monitoring simple and composite architectures, components and services
  • Semantic (or analytic) architectural and behavioral models for monitoring of software systems
  • Dynamic reconfiguration of cloud and mobile applications
  • Collaborative planning and decision making
  • Collaborative technologies for ensuring autonomic properties
  • Predictive management of adaptability
  • Collaborative Management of autonomic properties
  • Experiences in practical adaptive and reconfigurable applications
  • Tools and prototypes for managing adaptability of applications
  • IoT and cyber physical systems
Important dates

Abstracts Submission: Mon 12 May 2025
Papers Submission: Mon 19 May 2025 Fri 6 Jun 2025
Decision Notification: Fri 20 Jun 2025
Camera-Ready Submission : Fri 27 Jun 2025
Early registration: Fri 27 Jun 2025
Paper Submission
The workshop invites two types of submissions:
  • Full research and experience papers, the length of full/experience workshop papers shall be maximum 16 pages (including figures, references, etc.).
  • Short and position papers, presenting promising preliminary results from work-in-progress or research challenges, the length of short papers shall be maximum 8 pages plus 2 pages only for references.
All contributions must be written in English and formatted according to the Springer LNCS style. Contributions need to be submitted in PDF format via EasyChair. Papers follow a single-blinded reviewing process and must contain original unpublished work, describe significant novel contributions, and provide evidence about the validation of the results. Authors must upload their paper as PDF file using the EasyChair submission system.


Workshop chairs
Ismael Bouassida Rodriguez, ReDCAD, University of Sfax, Tunisia
Nesrine Khabou, ReDCAD, University of Sfax, Tunisia
Khalil Drira, LAAS-CNRS, Univ Toulouse, France
Publicity chair
Bechir Zalila, ReDCAD, University of Sfax, Tunisia
Web chair
Mouna Rekik, MIRACL, University of Sfax, Tunisia
Program Committee Members (TBC)
Abdulatif Alabdulatif Qassim University, Saudi Arabia
Bernd Freisleben University of Marburg , Germany
Carlos E. Cuesta Rey Juan Carlos University , Spain
Chouki Tibermacine University of Montpellier , France
Claudia Raibulet University of Milano-Bicocca , Italy
Cecilia Rubira University of Campinas , Brazil
Elisa Yumi Nakagawa University of São Paulo , Brasil
Flavio Oquendo IRISA (UMR CNRS) - Univ. Bretagne-Sud (UBS) , France
Ernesto Exposito Univ de Pau et des Pays de l'Adour , France
Federico Bergenti the University of Parma , Italy
Francisco Moo-Mena Autonomous University of Yucatán , Mexico
Fairouz Fakhfakh University of Sfax , Tunisia
Henry Muccini University of L'Aquila , Italy
Ian Gorton Northeastern University , USA
Imen Abdennadher University of Sfax , Tunisia
Mehdi Khouja University of Gabes , Tunisia
Miriam Capretz University of Western Ontario , Canada
Mohamed Mosbah LaBRI - University of Bordeaux , France
Mohamed Jmaiel University of Sfax , Tunisia
Mohand-Said Hacid Univ Claude Bernard Lyon 1 - UCBL , France
Mouna Rekik University of Sfax , Tunisia
Mourad Oussalah LINA Laboratory, University of Nantes , France
Nikolaos Georgantas INRIA , France
Philippe Roose LIUPPA/UPPA , France
Riadh Ben Halima University of Sfax , Tunisia
Sami Yangui University of Toulouse , France
Saul Pomares Autonomous University of Yucatán , Mexico
Slim Kallel University of Sfax , Tunisia
Uwe Zdun University of Vienna , Austria
Wafa Gabsi University of Sfax , Tunisia
Yamine Ait Ameur IRIT/INPT-ENSEEIHT , France
Contact
Dr. Ismael Bouassida Rodriguez
bouassida@redcad.org
ReDCAD Reserach Laboratory
B.P. 1173, Sfax, Tunisia

 

Sponsors:


Research Lab. on Development and Control of Distributed Applications
Laboratoire d'Analyse et d'Architecture des Systèmes
Association Of Computer Science And Mathematics