The authors describe a model of real-time distributed computing systems, making the simplifying assumption that the system is dedicated. They present the hardware architecture, operation and implementation of the noninvasive monitoring system which consists of two major components: the interface module and the development module. The interface module can be considered as the front end of the monitoring system. The development module is the host computer for the interface module. The authors describe their approach to monitoring and demonstrate how it can be used to support the testing and debugging of real-time distributed computing systems. Essentially, the monitoring system supports different abstraction levels of monitoring according to application purposes. It can monitor process-level activities (e.g. procedure calls) as well as instruction-level activities (e.g. step-by-step instruction trace). A detailed description of process-level monitoring is given.
