IWAN 2005 Tutorials

IWAN 2005 offers the unique one-day tutorial program dedicated to insightful studies of fundamentals of active and programmable networks: adaptive behaviours and self-organised services re-deployment. The tutorials complement each other by tackling software view and bio-inspired view, engineering and emergence, adaptation and complexity. The tutorial program consists of two half-day parts given 21 November 2005 by known professionals actively shaping the field.

All IWAN 2005 attendees are invited to register for the tutorial program that we believe is an excellent mind-setting for the main program of IWAN!

The regsitration fee for THE WHOLE TUTORIAL PROGRAM of the two parts is only EUR 80,00

Part 1 (morning)
Title: Adaptive middleware
Instructors: Simon Dobson and Paddy Nixon, Univesity College Dublin

Part 2 (afternoon)
Title: From understanding self-organization in biology to managing artificial complex systems
Instructors: Fabrice Saffre, British Telecom Jose Halloy Universite libre de Bruxelles


Tutorial 1:

From understanding self-organization in biology to managing artificial complex systems

A half-day tutorial at IWAN'05, Sophia-Antipolis, France 21 November 2005.

Motivation

There is widespread recognition that the increasing complexity of distributed computing systems, comprised of individual devices with variable purposes, requirements, abilities etc., cannot be efficiently managed throu gh centralised control. However, the emergent properties likely to manifest themselves in large ensembles of interacting "self-*" devices often receive limited attention and are sometimes dismissed as undesirable side- effects. This is arguably a serious mistake, as understanding and learning how to take advantage of those pro perties could be pivotal to the succes of the autonomic computing and communication paradigm.

Purpose

The purpose of this tutorial two-fold:

  1. To explain how an in-depth understanding of the interplay between local mechanisms and global dynamics can reveal economical ways of influencing complex biological systems.
  2. To demonstrate how this methodology can be adapted to manage artificial systems comprised of units whose i ndividual behaviour is "simple" (i.e. "non-complex"), but who collectively exhibit emergent properties.

Learning objectives

Attendees will be introduced to a variety of examples in which self-organization is found to play a critical role in the emergence of adaptive patterns in biological systems. They will be presented with experimental ev idence that such systems can be "piloted" toward a chosen macroscopic state by exploiting their dynamical pro perties. The notions prensented will be compared with the so-called "swarm intelligence" and "ant colony optimization" frameworks.

In the second part of the tutorial, attendees will be shown how the same logic can be applied to the decentra lised management of network-based collaborative computing, through autonomic load-balancing, topology adjustm ent, and deployment of software components. They will learn that at least some artificial complex systems are eminently predictable (in probabilistic terms) and can be driven toward a desirable state, with limited (or even in the absence of) external intervention.

Content

  1. An introduction to self-organisation in biology
  2. Predicting and influencing system behaviour
  3. Proof-of-concept: mixed societies of animals and robots
  4. Emergent properties in artificial networks
  5. Self-organised service (re-)deployment
  6. Emergent load-balancing throug co-operation
  7. Conclusions

Attendees

We would envisage a maximum of 25 participants. The tutorial materials will be made available on CD ROM.

About the presenters

Fabrice Saffre is a Principal Research Professional in British Telecom's Pervasive ICT Research Centre. Over the last 5 years, he has conducted extensive research in the field of Nature-inspired computing. José Halloy is a Senior Researcher at the Université libre de Bruxelles. He specialises in studying self-orga nization in a variety of biological systems, from the cellular level to animal societies.

Tutorial 2:

Adaptive middleware

A half-day tutorial at IWAN'05, Sophia-Antipolis, France 21 November 2005

Motivation

Traditional systems engineering has concentrated on the correctness and robustness of systems within a well-defined and clearly delineated environment. Recent trends in open systems, ad hoc networking, autonomic management and pervasive computing are leading to a situation in which a system's configuration and behaviour must adapt to a more dynamic environment in order to continue to deliver its required services. This poses a challenge for existing methodologies and tools.

Purpose

The purpose of this tutorial is to introduce attendees to the issues and tools -- both existing and emerging -- in the area of adaptive middleware. The emphasis is on understanding the relationship between a system's behaviour and the changes in its environment to which it must adapt, which allows us to develop principled approaches to defining and managing adaptive systems. These techniques are to a large extent technology-neutral, and so may be applied across a range of scales from sensor networks to enterprise information systems.

Learning objectives

Attendees will acquire a familiarity with the business and technological drivers that affect adaptive systems development. They will understand the relationship between environment and behaviour, and be introduced to some techniques for leveraging this relationship to design adaptive behaviours. They will acquire an high-level view of the application of these techniques to existing and emerging tools and platforms.

Content

  1. The landscape of systems engineering
  2. Management versus autonomics
  3. Modeling environment and behaviour
  4. Tools and technniques
  5. Emerging trends

Attendees

We would envisage a maximum of 25 participants. The tutorial materials will be made available on CD ROM.

About the presenters

Simon Dobson and Paddy Nixon are with UCD Dublin's Systems Research Group. Between them they have around 30 years' experience with distributed systems, middleware, programming languages, pervasive and autonomic systems in both academic and commercial contexts.


Please contact Jean-Patrick Gelas to update this page.