Work-in-Progress: Generalized Demand-Based Schedulability Test for Dual-Criticality Sporadic Task Model

Abstract

In this paper, we consider the scheduling of dual-criticality sporadic task systems with arbitrary deadlines using demand bound functions. In dual-criticality systems, tasks are assigned either low-criticality or high-criticality based on assurance needs with associated worst-case execution times. Arbitrary deadlines are those that allow the deadline to be larger than the minimum separation between consecutive task instances. Demand bound functions have been used to successfully schedule dual-criticality task sets for constrained deadlines, i.e., deadlines that are always less than or equal to minimum inter-arrival separation time. We formulate a new demand bound function for a more generalized dual-criticality task system with both constrained and arbitrary deadlines on a preemptive uniprocessor.