Towards a framework and a design methodology for autonomic integrated systems

Abstract

The transition from microelectronics to nanoelectronics reaches physical limits and results in a paradigm shift in the design and fabrication of electronic circuits. The conservative worst-case-approach is no longer feasible and has to be replaced by new design methods. These new design methods and tools have to guarantee reliable and robust systems in spite of unsafe and faulty functions on the lowest process levels. This paper proposes autonomic or organic computing principles to be applied to hardware design methods for future SoC solutions. Incorporating self-calibration, fault tolerance or even self-healing concepts into IC systems represents a major conceptual shift which requires new design processes and tools. In the future, guarantee of functional correctness at the chip level will include self-configuration of adaptable components and flexible interfaces supporting a flexible component composition within complex SoC systems. A high quality design process leading to more reliable systems is instrumental to secure a leading position in integrated system design among international competition. Of special interest are typical European application areas like automotive electronics, mobile systems, medical technology, smartcards, etc.