Abstract

Classically, the temporal validation of a hard real-time application is performed using the WCET's, for instance by means of simulation. But since the actual behaviour of the application uses the ACET's which may be shorter than the WCET's, the simulated and the effective behaviours may be different. Verifying whether an application behaves in accordance to a given scheduling strategy requires to precisely specify how the application is expected to behave in the case of ACET's shorter than the WCET's. For that aim, we define two notions of compliance. The non-flexible compliance imposes the strict respect of the start times given by the simulated schedule, whereas the flexible one tolerates a higher level of conservatism. We then discuss the advantages of each, and show that using any of them preserves the validity of the behaviour of the application.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call