Algorithms and optimality of scheduling soft aperiodic requests in fixed-priority preemptive systems

Abstract

In this paper, we investigate the problem of scheduling soft aperiodic requests in systems where periodic tasks are scheduled on a fixed-priority, preemptive basis. First, we show that given any queueing discipline for the aperiodic requests, no scheduling algorithm can minimize the response time of every aperiodic request and guarantee that the deadlines of the periodic tasks are met when the periodic tasks are scheduled on a fixed-priority, preemptive basis. We then develop two algorithms: Algorithm $\mathcal{L}$ is locally optimal in that it minimizes the response time of the aperiodic request at the head of the aperiodic service queue. Algorithm $\mathcal{G}$ is globally optimal in that it completes the current backlog of work in the aperiodic service queue as early as possible.