An Adaptive Value-Based Scheduler and Its RT-Linux Implementation

Abstract

In dynamic real-time systems, value-based scheduling aims to achieve graceful degradation during overloads, in addition to maintaining a high schedulability during normal and underloads. The objective of this paper is twofold: (1) to propose an adaptive value-based scheduler for multiprocessor real-time systems aimed at maintaining a high system value with less deadline misses, and (2) to present the implementation of the proposed scheduler in a Linux based real-time operating system, RT-Linux, which in its current form does not employ a notion of task value. We evaluate the performance of the proposed scheduler in terms of two performance metrics, namely, “value ratio” and “success ratio” through both simulation and implementation.KeywordsSuccess RatioHeuristic FunctionTask QueuePropose Schedule SchemeBranch FunctionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.