https://www.mdu.se/

Hierarchical scheduling provides a modular framework for integrating, scheduling and guaranteeing timing constraints of compositional real-time systems. In such a scheduling framework, all modules should receive a sufficient portion of the shared CPU to be able to guarantee timing constraints of their internal parts. In dynamic systems i.e., systems where the execution time of tasks are subjected to sudden and drastic changes during run-time, assigning fixed CPU portions to the modules is conducive to either low CPU utilization or numerous task deadline misses. In this paper, in order to address this problem, we propose an adaptive CPU allocation method which dynamically assigns CPU portions to the modules during run-time based on their current CPU demand. Besides, the presented approach is evaluated using a series of different simulations. In addition, we present a method for scheduling modules in situations when the CPU resource is not sufficient for scheduling all modules. We introduce the notion of module (subsystem) criticality, and in an overload situation we distribute the CPU resource based on the criticality of modules.

We have introduced an adaptive hierarchicalscheduling framework as a solution for composing dynamic realtime systems, i.e., systems where the CPU demand of its tasks aresubjected to unknown and potentially drastic changes during runtime. The framework consists of a controller which periodicallyadapts the system to the current load situation. In this paper,we unveil and explore the detailed behavior and performanceof such an adaptive framework. Specifically, we investigate thecontroller configurations enabling efficient control parameterswhich maximizes performance, and we evaluate the adaptiveframework against a traditional static one.

In our previous work, we have introduced an adaptive hierarchical scheduling framework as a solution for composing dynamic real-time systems, i.e., systems where the CPU demand of their tasks are subjected to unknown and potentially drastic changes during run-time. The framework uses the PI controller which periodically adapts the system to the current load situation. The conventional PI controller despite simplicity and low CPU overhead, provides acceptable performance. However, increasing the pressure on the controller e.g, with an application consisting of multiple tasks with drastically oscillating execution times, degrades the performance of the PI controller. Therefore, in this paper we modify the structure of our adaptive framework by replacing the PI controller with a fuzzy controller to achieve better performance. Furthermore, we conduct a simulation based case study in which we compose dynamic tasks such as video decoder tasks with a set of static tasks into a single system, and we show that the new fuzzy controller outperforms our previous PI controller.