Framework for Evaluation of Schedulability Tests for Mixed-Criticality Systems

Mixed-criticality real-time system models were proven to be useful in the processes of system verification, system validation, and safety assessment in safety-critical embedded systems. In recent years, a great number of various mixed-criticality models were investigated. These models are usually accompanied with corresponding schedulability tests. In the literature, schedulability tests are typically compared following the results of evaluation on synthetically generated task sets. Due to the large number of possible test scenarios and a significant effort that needs to be undertaken to implement a certain test, it is not an easy task to reproduce these results. Furthermore, as schedulability tests are used in safety-critical systems it is crucial that tests and results are reproducible. In this paper, we present a framework for evaluation of schedulability tests and show how it can be used to reproduce the most common results.