EUVIMEDEUVIMED entdecken
Crossref

Dynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems

Yiming Wang, Weizhe Zhang, Meng Hao, Weizhi Kong, Yuan Wen

ACM Transactions on Architecture and Code Optimization · 30.06.2025 · 22 (2) · 1-23

DOI: 10.1145/3716872 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 30.06.2025. Verfasst von Yiming Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 51 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, Joint Funds of the National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Power management and optimization play a significant role in modern computer systems, from battery-powered devices to servers running in data centers. Existing approaches for power capping fail to meet the requirements presented by dynamic workloads, and the situation becomes even more severe, given the divergent energy efficiency of workloads on heterogeneous hardware platforms. Adaptively optimizing energy consumption for dynamic workloads presents a great challenge to heterogeneous systems. To tackle this challenge, we present a machine learning based method to improve system-level power efficiency. We employ multi-agent deep reinforcement learning (MADRL) to automatically explore the relationship between long-term performance and the power budget for workloads of different types on classic CPU-GPU heterogeneous platforms. Our framework equips each device with an agent, enabling decentralized control over its power budget while maintaining centralized coordination to maximize the running time of applications within a power cap. We evaluate our approach against state-of-the-art methods on CPU-GPU platforms. Experimental results show that our method improves performance by an average of 8.5%. Additionally, our method is significantly more stable compared to the state-of-the-art heuristic approach.

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China
  • Joint Funds of the National Natural Science Foundation of China

Zitieren

APA

Wang, Y., Zhang, W., Hao, M., Kong, W., & Wen, Y. (2025). Dynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems. ACM Transactions on Architecture and Code Optimization, 22(2), 1-23. https://doi.org/10.1145/3716872

Vancouver

Wang Y, Zhang W, Hao M, Kong W, Wen Y. Dynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems. ACM Transactions on Architecture and Code Optimization. 2025;22(2):1-23. doi:10.1145/3716872

BibTeX
@article{Wang2025,
  author = {Wang, Yiming and Zhang, Weizhe and Hao, Meng and Kong, Weizhi and Wen, Yuan},
  title = {Dynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems},
  journal = {ACM Transactions on Architecture and Code Optimization},
  year = {2025},
  volume = {22},
  number = {2},
  pages = {1--23},
  doi = {10.1145/3716872}
}
Kostenlos starten →