The application of fault injection in the context of dependability benchmarking is far from being straightforward. One decisive issue to be addressed is to what extent injected faults are representative of actual faults. This paper proposes an approach to analyze the effects of real and injected faults.

The main goal of the experimental study reported in this paper is to investigate to what extent distinct fault injection techniques lead to similar consequences (errors and failures). The target system we are using to carry out our investigation is the Linux kernel as it provides a representative operating system. It is featuring full controllability and observability thanks to its open source status. Three types of softwareimplemented fault injection techniques are considered, namely: i) provision of invalid values to the parameters of the kernel calls, ii) corruption of the parameters of the kernel calls, and iii) corruption of the input parameters of the internal functions of the kernel. The workload being used for the experiments is tailored to activate selectively each functional component. The observations encompass typical kernel failure modes (e.g., exceptions and kernel hangs) as well as a detailed analysis of the reported error codes.

Component-Based Software Engineering (CBSE) does not yet fully address non-functional requirements of embedded systems. To reach this goal, we show how to extend a component model like FRACTAL with relevant abstractions such as threads, protection rings, or security domains. The FRACTAL Architecture Description Language (ADL) is extended by means of properties that tag components, bindings, and interfaces of the system architectural definition with execution schemes, dynamic reconfiguration strategies, protection and isolation patterns, or QoS features. Each extension captures a property-specific "system view" offering a sound basis to address some non-functional requirement. These extensions were experimented in the THINK framework, a C-based implementation of FRACTAL. Results show that THINK provides a generic and efficient approach to fully support these extensions thanks to a customizable toolchain.