A software library, a control backbone and user-specified recovery strategies to enhance the dependability of embedded systems

Fault tolerance in embedded distributed systems requires flexibility. Solutions should be based on pre-built and reusable elements, customisable for different applications and portable to different platforms. To this end, a framework is presented, consisting of a user library, a control backbone and a high-level description language (RL). The user library contains basic functions for fault tolerance. The control layer hooks these functions together and allows to co-ordinate actions. RL allows the developer to specify the recovery strategies of the application as a sort of second application layer; this separates functional from recovery aspects of an application, shortening the development cycle and improving maintainability. This paper describes the requirements to which this framework adheres and the context in which it can be instantiated; we detail the three entities and explain how an application can be made fault-tolerant in this framework approach.