Dynamically Reconfigurable Architectures

Motivation

The Dagstuhl seminar on "Dynamically Reconfigurable Architectures" has been a very successful meeting of people from different research areas - algorithms, hardware architectures and circuits as well as optical communication during the recent years. The seminar has shown that technological advances opens up new ways of implementing complex systems in a way that blurs the barriers between hardware and software component development, and that existing design tools do not seem to be adequate for the necessary new design styles. Furthermore, new advances in optical communication lead to feasible implementations of interconnection structures which are no longer theoretical only.

In recent years a rapidly growing interest in using reconfigurable computing architectures for realizing and developing application-specific computer systems has been observed. The advances in reconfigurable technologies, in algorithms for implementation approaches and in automatic mapping methods of algorithms into hardware and processor spaces form together a new computing paradigm of computing and programming, e.g. ”Computing in Space AND in Time”. This requires different and novel approaches in engineering for developing reconfigurable systems and implementing complex algorithms, including theory, architecture structures, algorithms, design systems and industrial applications that demonstrate the benefits of this promising way of computing. The fast pace of development is leaving industry not enough time to develop the necessary theoretical foundation that underpins CAD tools, OS, designs, architectures and circuit technologies. Traditional hardware and software design processes and the tools to support them are not adequate for the design of run time reconfigurable systems. Therefore, the plan for this seminar is to focus on the issues relevant to the development of support for the meantime also in industry attractive reconfigurable technologies. A special focus will be given to dynamically run-time reconfigurable (RTR) solutions, since here system adaptation and the advantages of this technology are highly visible.

Additional novel topics in this area of research are the emerging reconfigurable architectures based on nanotechnologies. The development of nanotechnologies has made rapid progress in recent years, and reconfigurable architectures based on nanotechnologies are already investigated in many different research projects. The expected complexity of embedded systems based on nanotechnologies requires extensive research in hardware architectures and implementation tools but also in approaches towards the design of self-adaptive systems. Furthermore, nanotechnologies will introduce uncertainties and unreliabilities that are unavoidable and undetectable during the design process, and one possibility to cope with them is to empower the system with capabilities to reconfigure and adapt itself in such a way that it can overcome its own failures. Self-adaptive systems are designed to manage decisions on-line, to perform on-line optimization and to adapt themselves to changing requirements from the application during run-time. Reconfiguration can be exploited in order to simplify the implementation of self-adaptive systems.

The seminar will cover: approaches for designing self-adaptive systems, architecture structures (also for nanotechnologies), circuit technologies, system architectures, tools for RTR, general/special purpose system, and of course, existing and new application domains, where (dynamically) reconfigurable computing is more effective than traditional and parallel/distributed architectures. This includes also an appropriate set of models for reconfigurable systems which open ways for application designers and industry to develop efficiently their systems using appropriate high level languages. Especially the risk minimizing factors (time-to-market!) and adaptation features (-> multipurpose/ multistandard possibilities!) are important arguments for industrial companies now, e.g. in (mobile) communication technologies, automotive area, etc. to integrate this flexible technology into their product strategy. Here Configurable Systems-on-Chip (CSoCs) solutions for embedded systems are giving valuable perspectives. We also think that the challenges posed by integrating optical technology with RTR should remain a considerable aspect of this seminar.

Furthermore, the recent initiative Organic Computing is another important field that will contribute from this seminar, due to the possibility for realization of self-X characteristics in embedded systems by exploiting reconfigurable hardware architectures. The forum should also involve experts from these areas to discuss potential use of reconfigurable hardware in these challenging areas. Expected outcomes are e.g the exchange of research results in the area of Reconfigurable Computing, identification of expected future research challenges, agreement on important future research topics and initiatives for international collaboration (industrial and academic).