Suche

3rd International Workshop on „Self-optimisation in Autonomic and Organic Computing Systems" (SAOS15)

 

Agenda

Tuesday, 24th March, 2015 

Session 1

Coffee Break

Session 2

Session 3

Coffee Break

Session 4

 

Call for Paper

Initiatives such as Autonomic Computing (AC) and Organic Computing (OC) are based on the insight that we are increasingly surrounded by large collections of autonomous systems, which are equipped with sensors and actuators, aware of their environment, communicating freely, and organising themselves in order to perform the required actions and services. The presence of networks of intelligent systems in our environment opens fascinating application areas but, at the same time, bears the problem of their controllability.

Hence, different design concepts (like the MAPE cycle and the Observer/Controller framework) have been developed to allow for a self-organised control process at runtime that relieves the designer from specifying all possibly occurring situations and configurations within the design process. Instead, the system itself takes over responsibility to find proper reactions on perceived changes in the environmental conditions. As designers are not able to foresee all possibly occurring situations and circumstances the system will face during its operation time the self-organisation process of the system will focus on self-optimising the system’s behaviour. Such self-optimising behaviour can be achieved at various levels of the system’s design, ranging from basic control architectures over self-organised coordination or collaboration methods and domain-specific optimisation techniques to the application and customisation of machine learning algorithms. Furthermore, several related topics (e.g. trust and security in collaborative systems) provide necessary functionality to enable self-optimising behaviour in AC and OC systems.

A special session will further address challenges caused by so-called Interwoven Systems (IwS). That is, invite contributions discussing concepts to achieve controllability and manageability of connected or coupled systems that were initially not meant to interact with each other. A key contribution towards a better control of IwS is a qualified support of self-integration in IwS elements. In order to make this challenge more accessible, effort can be subdivided into six distinct areas: Due to changes in the overall IwS, e.g., new elements, the overall system architecture (element structure, communication structure, algorithms, parameters, etc.) cannot be predetermined at design-time. Basis for a resulting self-integration process is, first, the ability of elements for online dependency detection and modelling and, second the ability for online goal adaptation. Then, techniques for a continuous re-design for self-integration, i.e., an adaptation of various aspects of an IwS system architecture can be established. These self-integration processes must be complemented by techniques for a long-term self-improvement, i.e., self-inspection and long-term self-adaptation. As IwS are open, vulnerable systems, we need techniques to model and consider trust, reputation, and security mechanisms. Methods for a theoretical analysis of self-integration processes at different levels will be crucial for a better understanding of IwS at design-time and to provide necessary information about the IwS at runtime.

Part A: Architectural concepts for self-optimising behaviour

 

Part B: Algorithms and methods for self-optimisation  

 

Part C: Applications for self-optimisation

Applications with self-optimising system behaviour, i.e. from the following domains:

 

Part D: Interwoven Systems & Self-Integration

 

 

Workshop details

 

Schedule and submission

 

Workshop organisation

 

Program committee: