Performance evaluation of distributed, intelligent real-time control systems

Online broadcasting of distributed computer control systems (DCCS) performance parameters (e.g. bandwidth, and real-time response time parameters, as well as future performance requests) via the underlying communication system, leads to an online, present and predictive time-discrete dynamic system knowledge. Implementing a model of the underlying communication system and merging the model with the time-discrete dynamic system knowledge leads to a basic time-discrete dynamic system knowledge model, which can be implemented within the DCCS and can be used to observe and to control the whole system. Implementing further stochastic parameter optimization results in an optimal, adaptive control of the stochastic, time-discrete process that describes the performance and the behaviour of the DCCS. We call this controlled DCCS a distributed intelligent computer control system (DICCS). This paper deals with basic performance evaluation of a DCCS's underlying fiber distributed data interface (FDDI) communication system and the evaluation of hard-bounded, worst-case real-time parameters describing the behaviour of the DCCS. The derived parameters shown in this paper can be used for an optimal, adaptive control of DICCS.