Towards automated system-level energy-efficiency optimisation using machine learning
Publication by
Benedict Herzog,
Stefan Reif,
Fabian Hügel,
Timo Hönig,
Wolfgang Schröder-Preikschat
Related to the Energy-, Latency- And Resilience-aware Networking (e.LARN) project
Published in e-Energy '21: Proceedings of the Twelfth ACM International Conference on Future Energy Systems, 2021
Related to the Energy-, Latency- And Resilience-aware Networking (e.LARN) project
Published in e-Energy '21: Proceedings of the Twelfth ACM International Conference on Future Energy Systems, 2021
Abstract:
Modern computing systems need to execute applications in an energy-efficient manner. To this end, operating systems, middleware, and run-time systems offer plenty of parameters that support fine-tuning their behaviour. However, their individual and combined impact on performance and power draw is so complex that this optimisation potential is often ignored in practice. This paper therefore discusses a cross-layer system design that uses machine learning internally to enable fine-tuning run-time systems to their current workload. Our approach includes all layers, from the hardware to the application, considering both performance and power draw.