When hard periodic and rm aperiodic tasks are jointly scheduled in the same system, the processor workload can vary according to the arrival times of aperiodic requests. In order to guarantee the schedulability of the periodic task set, in overload conditions some aperiodic tasks must be r ejected.In this paper we propose a t e chnique that, in overload conditions, adds robustness to the joint scheduling of periodic and aperiodic tasks in systems with dynamic priorities. Our technique is based on an aperiodic server, called Total Bandwidth server, already proven e ective in a previous work. Here the algorithm is rst extended to e ciently handle rm aperiodic tasks and then integrated with a robust guarantee mechanism that allows to achieve graceful degradation in case of transient overloads. Extensive simulations show that the proposed new algorithm is e ective in all workload conditions.
ÐIn this paper, we present a new scheduling approach for servicing soft aperiodic requests in a hard real-time environment, where a set of hard periodic tasks is scheduled using the Earliest Deadline First algorithm. The main characteristic of the proposed algorithm is that it achieves full processor utilization and optimal aperiodic responsiveness, still guaranteeing the execution of the periodic tasks. Another interesting feature of the proposed algorithm is that it can easily be tuned to balance performance versus complexity for adapting it to different application requirements. Schedulability issues, performance results, and implementation complexity of the algorithm are discussed and compared with other methods, such as Background, the Total Bandwidth Server, and the Slack Stealer. Resource reclaiming and extensions to more general cases are also considered. Extensive simulations show that a substantial improvement can be achieved with a little increase of complexity, ranging from the performance of the Total Bandwidth Server up to the optimal behavior.
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.