This paper presents a new type of scheduling algorithm for servicing soft deadline aperiodic tasks in adaptable fixed-priority real-time systems. The major goals of our proposed task scheduling are not only to guarantee all the deadlines of periodic tasks and to obtain fast response time for aperiodic tasks, but also to retain considerable scheduling predictability. To achieve these goals, we have adopted a new aperiodic task scheduling principle in which a normal fixed-priority assignment strategy and the information on a preassignment table built off-line are properly mingled somewhat dynamically according to the aperiodic tasks' arrivals at runtime. The paper also shows some simulation results in terms of the average aperiodic response time, verifying that the new algorithm offers significant performance improvements over other conventional joint scheduling algorithms, especially under a heavy transient overload.
In this paper, we present a preemptive joint scheduling of hard deadline periodic and hard deadline aperiodic tasks on a uniprocessor real-time system. The scheduling has extended the Critical Task Indicating (CT1) algorithm [4] of which simulation study shows a considerable performance improvement over the other soft aperiodic task schedulings, especially under a heavy transient overload. Since a hard deadline aperiodic task has its own deadline, the proposed algorithm has a decision making mechanism that performs the on-line acceptance/rejection test upon its arrival. For simplicity and good performance, the algorithm reuses the original CTI table being used in the CTI algorithm as a slack search domain. Moreover, by searching the CTI table similarly to a circular list, it has removed the problem of search space limitation caused by the hyperperiod bound.
ABSTmCTThe major drawback of the slack-stealing based I schedulings for aperiodic requests is a high computational complexity to calculate the slack which in consequence makes them not be practical. In this paper, we present a sofi-aperiodic task scheduling algorithm, called Alternative Priority Scheduling (APS), which has a simple slack calculation method in &namic priority systems. The proposed algorithm has been extended the EDF-CTI (Earliest Deadline First-Critical Task Indicating;) Algorithm f 6J,[8' developed by the authors. The APS algorithm references the off-line built CTI table and chooses either an EDF or a CEF (Critical Execution time First) algorithm alternatively at run-time. This paper also demcmtrates the optimality of the APS algorithm. Our ,simulation study shows that the APS algorithm, in most cases, is slightb better than the EDF-CTI algorithm and the other so$-aperiodic schedulings in terms of the short response time of aperiodic requests, and considerably improves the previous algorithms in a high workload
In this paper, we propose a reasonably simple and near-optimal soft-aperiodic task scheduling algorithm in dynamic priority systems. The proposed algorithm has extended the EDF-CTI (Earliest Deadline First-Critical Task Indicating) Algorithm [4/ in such a way of modifying the slack calculation method which in turn reduces the computational complexity of slack calculation and resolves the unit scheduling problems. The paper also demonstrates near optimality of the algorithm.Our simulation study shows that the proposed algorithm, in most cases, is slightly better than the EDF-CTI algorithm in terms of short response time of aperiodic requests in a high workload.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.