Improved conditions for bounded tardiness under EPDF Pfair multiprocessor scheduling

The earliest-pseudo-deadline-first (EPDF) Pfair algorithm is more efficient than other known Pfair schedulingalgorithms, but is not optimal for scheduling recurrent real-time task systems on more than two identical pro-cessors. Although not optimal, EPDF may be preferable for real-time systems instantiated on less-powerfulplatforms, those with soft timing constraints, or those whose task composition can change at run-time. Inprior work, Srinivasan and Anderson established a sufficient per-task utilization restriction for ensuring atardiness of at most q quanta, where q ≥ 1, under EPDF. They also conjectured that under this algorithm,a tardiness bound of one quantum applies to task systems that are not subject to any restriction otherthan the obvious restrictions, namely, that the total system utilization not exceed the available processingcapacity and per-task utilizations not exceed 1.0. In this paper, we present counterexamples that show thattheir conjecture is false and present sufficient per-task utilization restrictions that are more liberal thantheirs. For ensuring a tardiness bound of one quantum, our restriction presents an improvement of 50% overthe previous one.