The Imprecise DEVS approach combines the advantages of imprecise computation with a formal modeling methodology in order to avoid the transient overloads on real-time systems. This process requires efficient scheduling methods to find the best schedule of the computations to guarantee the deadlines and to increase the quality of the results by reducing the discarding of the optional computations. This work introduces a solution to integrate schedulability analysis with Imprecise DEVS to improve the predictability and the feasibility for scheduling. The proposed schedulability analysis and scheduling methods are based on Earliest Deadline First, priority-driven and mandatory-first approaches and they are considered to be executed dynamically. The schedulability tests contribute to avoid unnecessary discarding of optional computations improving the quality of the results.

Additional Metadata
Keywords Imprecise DEVS, Real-time systems, Schedulability analysis
Conference 2016 TMS/DEVS Symposium on Theory of Modeling and Simulation, TMS/DEVS 2016, Part of the 2016 Spring Simulation Multiconference, SpringSim 2016
Citation
De Mello, B.A. (Braulio Adriano), & Wainer, G.A. (2016). Scheduling predictability in I-DEVS by schedulability analysis. In Proceedings of the 2016 Spring Simulation Multiconference - TMS/DEVS Symposium on Theory of Modeling and Simulation, TMS/DEVS 2016.