Heuristic algorithms for priority assignment in flow shops

An end to end approach for scheduling hard real time periodic transactions in a multiprocessor or distributed system is considered. Each transaction consists of a chain of tasks, where each task is allocated to a processor based on its resource requirements. Assuming fixed priority scheduling on processors, the problem is to determine the priority of each task such that the end to end deadline for each transaction is met. Heuristic algorithms for priority assignment in a periodic flow shop model are introduced and their abilities to find a feasible schedule are compared to the existing heuristic algorithms by using simulation methods.