On-line scheduling in multiprocessor systems based on continuous control using Timed Continuous Petri Nets

Abstract

This work presents a fluid-time scheduler based on a sliding mode controller where the sliding surface is related to fluid task executions. The scheduler is applied to a model of tasks and CPUs designed with Timed Continuous Petri Nets (TCPN) under the infinite server semantics (ISS). Also, the paper proposes an implementation of this fluid scheduler as a feasible discrete scheduler where the number of task migrations and preemptions is bounded.