174 : Scheduling Systems

NMCS (networked measurement–control systems) are widely used in applications ranging from difficult process control to simple discrete manufacturing. Usually, these systems impose real-time requirements on the nodes and communication networks (Kopetz, 1997; Sha and Sathaye, 1995). Stankovic (1988) defines realtime computing systems as those systems in which the correctness of the system depends not only on the logical result of computation but also on the time at which the results are produced. There are many examples of real-time computing systems, such as control systems, flight control systems, and robotics. A typical real-time computing system has a real-time program running in the system, which reads inputs from input devices, processes these inputs, and produces outputs to be sent to output devices – see also Article 78, Embedded Systems, Volume 1. The time between the arrival of an input from a device and the completion of the processing for that input is called the response time for the device (Joseph and Pandya, 1986). The relative deadline for the device can be defined as the maximum interval between the instant of the input arrival and the completion of the processing for that input. The required response time for a device must be smaller or equal to its relative deadline. In NMCS, the relative deadline is often assumed to be equal to the period. The task deadline value is set by object requirements but the value of the worst-case response time is a result of simulation methods or analytical techniques used to do the timing analysis. It is assumed that each input and output device is assigned a task of the application program and that the tasks share the same processor. The problem of determining whether the processor will meet its peak-processing load becomes one of schedulability analysis of tasks in NMCS (Burns, 1991). If the input (sensor node) and output (actuator node) devices are network components, then the input device processes events and generates messages to the output device. The problem of determining whether the communication system will meet its peak load is one of schedulability analysis of tasks. Consideration of priority levels is crucial to a real-time system. If different inputs have different response time requirements, we need to consider different priority levels to schedule the related processing tasks. The problem is to determine whether for a given assignment of priority levels, the system will meet its peak load – see also Article 165,