Résumé des résultats (Abstract)
(Anglais)
|
Information systems like the physical Internet, the world wide web, telephone networks, mobile ad-hoc networks, or peer-to-peer networks have reached a level that puts them beyond our ability to deploy them, manage them, and keep them functioning correctly through traditional techniques. Reasons for this are their sheer size with millions of users and interconnected devices and their dynamics; they evolve dynamically over time, i.e., components change or are removed or inserted permanently. For these systems, we have to abandon the goal of global optimality. Rather we have to concentrate our efforts on seeking self-regulating and self-repairing mechanisms that are decentralized, scalable, and that are able to adapt to changes in their environments and maintain the system in states corresponding to acceptable behaviour, avoiding those that are undesirable or unstable.
The goal of this project is to develop methods, techniques and tools to cope with challenges imposed-by such systems, in an interdisciplinary effort of Computer Science, Physics, Biology, and Economy. The objectives we have identified and want to address are: To understand the structure, self organisation and dynamics of large scale information systems, to provide methods, techniques and tools for controlling and optimizing such systems, to apply methods from market mechanism and game theory in order to understand and to organize the competition for resources like bandwidth, computing power, data etc, and to demonstrate the capabilities of our methods, techniques and tools in two application areas of high scientific and economic impact: A management platform for telecommunication transport networks that integrates the wide range of current and future data formats and services, and a decentralized, self-organizing Web search engine based on a peer-to-peer architecture, resulting in higher quality answers to search requests than current centralized search engines.
|
