An Optimal Multiprocessor Real-Time Scheduling Algorithm

An optimal scheduling algorithm is described that feasibly schedules a set of m periodic tasks on n processors before their respective deadlines, if the task set satisses certain conditions. The complexity of this scheduling algorithm in terms of the number of scheduled tasks and the number of processors and upper bounds on the number of preemptions in a given time interval and for any single task is also derived. The optimal algorithm is shown to be particularly useful when schedules are built from the integral ow values obtained from the corresponding maximum ow network.