PREDATOR · Design for Predictability and Efficiency

The project proposal is concerned with embedded systems that are characterized by efficiency requirements such as average-case performance, resource utilization, and power consumption on the one hand and worst-case constraints on the other. This combination of requirements typically occurs in application domains such as automotive, aeronautics, multi-media and industrial automation. Embedded systems with critical constraints need offline guarantees for the satisfaction of these constraints. Unfortunately, it can be observed that in computer system design the gap between average case and worst-case behaviour increases rapidly. This entails a decreasing precision of performance-analysis results, even for combination of the strongest analyses available. Therefore, a new research and design discipline is proposed that looks at predictability and efficiency in a synergistic manner and that involves all levels of abstraction and implementation in embedded-system design. This paradigm shift overcomes the tendency to either optimize efficiency only or predictability only in favour of an approach that takes into account the multi-objective nature of the problem. <br/>The proposed approach consists of a combination of several methods, i.e.<br/>(a)\tdesign-space exploration on the hardware architecture level to identify good designs offering combinations of strong performance with good predictability, <br/>(b)\ta synergistic development of models, design methods and matching analysis tools that extract precise system-behaviour properties, and <br/>(c)\ta transformation of the established separation-of-concerns abstraction principle into a new principle, resource-aware abstraction.<br/>Partners from the automotive and aeronautics domains pose design challenges based on experience in the design of time-critical embedded systems. These challenges will be taken up by the academic partners. Prototype architectures, design methods and analysis tools will be developed to solve the challenges