EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Green MLOps to Green GenOps: An Empirical Study of Energy Consumption in Discriminative and Generative AI Operations

Adrián Sánchez-Mompó, Ioannis Mavromatis, Peizheng Li, Konstantinos Katsaros, Aftab Khan

Information · 30.03.2025 · 16 (4) · 281

DOI: 10.3390/info16040281 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Information, online erschienen am 30.03.2025. Verfasst von Adrián Sánchez-Mompó und 4 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Toshiba Europe Ltd., Bristol Research and Innovation Laboratory (BRIL), Future Open Networks Research Challenge (FONRC) sponsored by the Department of Science Innovation and Technology (DSIT).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study presents an empirical investigation into the energy consumption of discriminative and generative AI models within real-world MLOps pipelines. For discriminative models, we examine various architectures and hyperparameters during training and inference and identify energy-efficient practices. For generative AI, large language models (LLMs) are assessed, with a focus primarily on energy consumption across different model sizes and varying service requests. Our study employs software-based power measurements, ensuring ease of replication across diverse configurations, models, and datasets. We analyse multiple models and hardware setups to uncover correlations among various metrics, identifying key contributors to energy consumption. The results indicate that, for discriminative models, optimising architectures, hyperparameters, and hardware can significantly reduce energy consumption without sacrificing performance. For LLMs, energy efficiency depends on balancing model size, reasoning complexity, and request-handling capacity, as larger models do not necessarily consume more energy when utilisation remains low. This analysis provides practical guidelines for designing green and sustainable ML operations, emphasising energy consumption and carbon-footprint reductions while maintaining performance. This paper can serve as a benchmark for accurately estimating total energy use across different types of AI models.

Förderung

  • Toshiba Europe Ltd.
  • Bristol Research and Innovation Laboratory (BRIL)
  • Future Open Networks Research Challenge (FONRC) sponsored by the Department of Science Innovation and Technology (DSIT)

Zitieren

APA

Sánchez-Mompó, A., Mavromatis, I., Li, P., Katsaros, K., & Khan, A. (2025). Green MLOps to Green GenOps: An Empirical Study of Energy Consumption in Discriminative and Generative AI Operations. Information, 16(4), 281. https://doi.org/10.3390/info16040281

Vancouver

Sánchez-Mompó A, Mavromatis I, Li P, Katsaros K, Khan A. Green MLOps to Green GenOps: An Empirical Study of Energy Consumption in Discriminative and Generative AI Operations. Information. 2025;16(4):281. doi:10.3390/info16040281

BibTeX
@article{SnchezMomp2025,
  author = {Sánchez-Mompó, Adrián and Mavromatis, Ioannis and Li, Peizheng and Katsaros, Konstantinos and Khan, Aftab},
  title = {Green MLOps to Green GenOps: An Empirical Study of Energy Consumption in Discriminative and Generative AI Operations},
  journal = {Information},
  year = {2025},
  volume = {16},
  number = {4},
  pages = {281},
  doi = {10.3390/info16040281}
}
Kostenlos starten →