[Back]

I. Smaili:
"Monitoring of Distributed Time-Triggered Systems: Case Study";
2004.

A main prerequisite for the successful development of real-time systems is the capability for debugging and monitoring of these systems. The goal of the monitoring system is to collect all needed, i.e., relevant monitoring data (MD), from which the run-time behavior of the target system at the intended abstraction level can be reproduced. The correctness of real-time systems (RTS) depends not only on the results they deliver, but also on the point in time on which these results are delivered. Therefore, the key issue during monitoring of RTS is to keep the interference deterministic that is caused by a monitoring system on a RTS, if it cannot be completely removed. This interference depends on the way how MD are collected from the target system, and on the amount of collected MD. This paper describes the design and implementation of a monitoring system for monitoring of Time-Triggered Architecture (TTA) target systems at the node abstraction levels, i.e., at the operating system (OS) and the application abstraction levels. This serves as a case study for the presented concepts on monitoring of TTA target systems at the above mentioned abstraction levels. The brake-by-wire (BBW) control system is used as a target system, which is a distributed simulation program designed by Volvo Technological Development.