Analytical Approximations of Real-Time Systems with Preemptive Priorities Working Under Maximal Load

We consider a real-time data acquisition and processing multiserver system with identical servers (such as unmanned air vehicles, overhearing devices, machine controllers, medical monitoring devices, etc.) which can be maintained/programmed for different kinds of activities (e.g. passive or active). Such a system provides a service for real-time tasks and works under maximal load with preemptive priorities assigned for servers of different activity type.

We study two models with ample and limited maintenance facilities respectively. These systems can be presented as queueing networks, but their steady-state probabilities cannot be obtained in product form. Corresponding set of balance equations can be solved only numerically.

We consider several analytical approximations having product-form solutions and show, which one of them is the best.