Predictable distributed dynamic scheduling in RTDOS

Abstract

Future dynamic distributed hard real-time systems may control unpredictable environments, and will need predictable and flexible runtime systems that can handle unknown and changing task populations. In this extreme case not only is task scheduling dynamic, but the system topology and architecture might be adapted to unforeseen configurations. The paper addresses the difficult problem of dynamic distributed task scheduling. A new predictable dynamic deadline guarantee scheme has been designed and implemented in the authors' real-time distributed operating system (RTDOS). The algorithms have been proven to guarantee task deadlines even during transient overloads. RTDOS is designed to tackle unpredictable and highly dynamic environments; therefore its task model is quite unrestrictive, supporting periodic and aperiodic tasks both with arbitrary release times and deadlines. The authors place no restriction on the inter-arrival times between successive aperiodic task instances. Moreover, their resource scheduling considers precedence constraints, device allocation and communication requirements. As a consequence, the results can be generalised and applied to many real-time domains. Furthermore, the complexity of the authors' schedulability test has been proved to be O(n). A domain-wide scheduler is described, which maintains its predicted maximum response time even when local nodes are under heavy load. Experimental results confirm the proven expectations of our scheduling scheme.