Networked embedded systems
Many embedded systems encompass clusters of networked devices to satisfy resource requirements exceeding the resources of a single node system. In addition, today’s technology does not support the manufacturing of electronic devices with failure rates low enough to meet the reliability requirements of ultra-dependable systems. Since failure rates of node computers are usually in the order of 10-5 to 10-6, ultra-dependable applications require the system as a whole to be more reliable than any one of its node computers. This can only be achieved by utilizing fault-tolerance strategies that enable the continued operation of the system in the presence of node failures.
For the networks of these distributed embedded system with clusters of networked devices, there are many different and, partially, contradicting requirements. Well-known tradeoffs are predictability versus flexibility or resource adequacy versus best-effort strategies. Therefore, the chosen system model depends strongly on the requirements of the application.
Research activities in this focus area include:
Integration of time-triggered and event-triggered communication (e.g., coexistence of different timing models, emulation of CAN in time-triggered networks)
Extension of time-triggered networks to support dynamic reconfiguration (e.g., mode switching, dynamic computation and deployment of schedules)
Fault-tolerance mechanisms (e.g., star coupler for fault isolation in CAN, network interface extension for traffic shaping)