Schedulability analysis of real-time multiframe cosimulations on multicore platforms

Abstract

For real-time simulations, the fidelity of simulation depends not only on the functional accuracy of simulation but also on its timeliness. It is helpful for simulation designers if they can analyze and verify that a simulation will always meet its timing requirements without unnecessarily sacrificing functional accuracy. Abstracting the simulated processes simply as software tasks allows us to transform the problem of verifying timeliness into a schedulability analysis problem where tasks are checked as to whether they are schedulable under the timing constraints or not. In this paper we extend a timed automaton-based framework due to Fersman and Yi for schedulability analysis of real-time systems, for the special case of real-time multiframe cosimulations. To the best of our knowledge, this work is the first to analyze the schedulability of single- or multiframe real-time simulations. We found that there are some special requirements posed by multiframe simulations, which necessitate changes and improvements in the existing framework designed for actual real-time systems. We made the required theoretical extensions to the framework and implemented our extended framework in UPPAAL, a tool for modeling, simulation, and verification of real-time systems modeled as timed automata. The functional correctness and resource requirements of the implemented framework are then demonstrated using simple examples.