An Operating System Architecture for Organic Computing in Embedded Real-Time Systems Florian Kluge, J¨ org Mische, Sascha Uhrig, and Theo Ungerer Department of Computer Science - University of Augsburg 86159 Augsburg, Germany {kluge, mische, uhrig, ungerer}@informatik.uni-augsburg.de Abstract. To overcome the rising complexity of computing systems, the paradigms of Autonomic Computing and Organic Computing have been introduced. By using an observer/controller architecture, Organic Com- puting aims to make embedded systems more life-like by providing them with so-called Self-X properties. Embedded real-time systems can also gain great benefit from these techniques. In this paper, we show what new requirements arise when introducing Autonomic/Organic Comput- ing into the area of real-time applications. These requirements flow into the architecture of the real-time operating system CAROS. CAROS com- bines several concepts to provide a solid base for the implementation of Self-X techniques in embedded real-time systems. We show the practi- cability of our concepts with a prototypical implementation on the mul- tithreaded CarCore microcontroller. 1 Introduction Today, embedded systems are constantly growing, and establishing whole net- works of Embedded Control Units (ECUs). For example, a car can contain over 70 ECUs fulfilling most different duties. With increasing size these networks become harder if not impossible to manage. The paradigms of Autonomic and Organic Computing promise to handle this topic. In 2001 IBM introduced Autonomic Computing (AC) [1, 2] to overcome the problem of increasing complexity of computing systems. AC focuses on self- management of large server systems by implementing the so-called Self-X prop- erties of self-configuration, self-healing, self-optimisation and self-protection (also referred to as “Self-CHOP”). To implement such self-management techniques, Autonomic Managers are proposed that control the system at runtime by a closed control loop of Monitoring, Analysis, Planning, and Execution (MAPE cycle). A few years later, Organic Computing (OC) [3] took up the Self-X concepts, focusing on distributed embedded systems. In general, AC/OC aspire to the de- velopment of robust, flexible and highly adaptive computing systems. To support the Self-X properties, Richter et al. [4] developed a generic observer/controller architecture similar to the MAPE cycle. A System under Observation/Control