EUVIMEDEUVIMED entdecken
Crossref

On Efficiency, Fairness and Security in AI Accelerator Resource Sharing: A Survey

Jiahua Huang, Weiwei Lin, Wentai Wu, Yang Wang, Haocheng Zhong, Xinhua Wang, Keqin Li

ACM Computing Surveys · 30.09.2025 · 57 (9) · 1-35

DOI: 10.1145/3721427 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Computing Surveys, online erschienen am 04.04.2025. Verfasst von Jiahua Huang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 144 Literaturangaben. Gefördert von: Guangdong Major Project of Basic and Applied Basic Research, National Natural Science Foundation of China, Guangxi Key Research and Development Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The effective and efficient utilization of AI accelerators represents a critical issue for the practitioners engaged in the field of deep learning. Practical evidence from companies such as Alibaba, SenseTime, and Microsoft reveals that the utilization of production GPU clusters in the industry is generally between 25% and 50%. This indicates a significant opportunity for improvement. To this end, AI accelerator resource sharing has emerged as a promising approach to the performance optimization of multi-tenant clusters. This survey covers this line of studies from 2016 to 2024, focusing primarily on system efficiency while also including discussion on fairness, interference, and security in AI accelerator sharing. We revisit the fundamentals and key concepts, followed by a comprehensive review of recent advances in the field. We find that over 70% of the studies focus on efficiency improvement. We also observe that approximately half of the reviewed studies have made their source code publicly available, while fewer than one-third of the studies did not utilize a physical machine for experimentation. Finally, based on the limitations of existing research, we outline several directions for future research concerning the integration of sharing with large language models (LLMs), coordination between schedulers and application-layer metrics, and collaboration among heterogeneous accelerators.

Förderung

  • Guangdong Major Project of Basic and Applied Basic Research
  • National Natural Science Foundation of China
  • Guangxi Key Research and Development Project
  • Guangzhou Development Zone Science and Technology Project
  • Fundamental Research Funds for the Central Universities
  • Major Key Project of PCL

Zitieren

APA

Huang, J., Lin, W., Wu, W., Wang, Y., Zhong, H., Wang, X., & Li, K. (2025). On Efficiency, Fairness and Security in AI Accelerator Resource Sharing: A Survey. ACM Computing Surveys, 57(9), 1-35. https://doi.org/10.1145/3721427

Vancouver

Huang J, Lin W, Wu W, Wang Y, Zhong H, Wang X, et al. On Efficiency, Fairness and Security in AI Accelerator Resource Sharing: A Survey. ACM Computing Surveys. 2025;57(9):1-35. doi:10.1145/3721427

BibTeX
@article{Huang2025,
  author = {Huang, Jiahua and Lin, Weiwei and Wu, Wentai and Wang, Yang and Zhong, Haocheng and Wang, Xinhua and Li, Keqin},
  title = {On Efficiency, Fairness and Security in AI Accelerator Resource Sharing: A Survey},
  journal = {ACM Computing Surveys},
  year = {2025},
  volume = {57},
  number = {9},
  pages = {1--35},
  doi = {10.1145/3721427}
}
Kostenlos starten →