ILP representation for Limited Preemption in Energy-Neutral Single-Core Systems
Résumé
In this work, we present an Integer Linear Programming (ILP) based approach that optimally selects preemption points for a set of real-time tasks. Beyond to meeting real-time constraints, the system must also consider energy availability constraints to ensure energy neutrality. This enables the design of autonomous low-end IoT devices with real-time constraints, minimizing maintenance operations to battery replacements.
We demonstrate that our system, despite being modeled using ILP, achieves strong temporal performance, even for large-scale problems. We evaluated its performance using the Gurobi solver across a wide range of synthetic experiments.
Domaines
Systèmes embarqués| Origine | Fichiers produits par l'(les) auteur(s) |
|---|