A Processor-Efficient Scheme for Supporting Fault-Tolerance in Rate-Monotonic Scheduling

We address the issue of supporting fault-tolerance in a real-time system, where the deadlines of periodic tasks are guaranteed by the Rate-Monotonic algorithm. The problem is stated as one to minimize the total number of processors required to execute a set of periodic tasks, each of which, for fault-tolerance purposes, has multiple versions. A simple but effective algorithm is proposed to solve the task allocation problem. The algorithm is shown to have superior worst-case and average-case performance.