There are many vital real-time systems such as avionics, mission-critical, industrial control, and space-mission, which cannot tolerate any request overruns. Safeness analysis of such systems is directly dependent on the scheduling strategy being used. Rate-monotonic and earliest deadline first are the leading scheduling strategies for these systems. Before deployment, it is essential to make sure no task misses occur for such hard real-time systems. This is usually done by formally proving the safeness of the system for the given set of tasks. Besides, a scheduler is preferred which performs better in unexpected situations such as when a transient fault occurs. The purpose of this paper is to first reveal some new behaviours of the intelligent rate-monotonic scheduler and then evaluate its performance in unexpected situations. Simulation results show that it performs better than both rate-monotonic and earliest deadline first in such situations.

Keywords