EUVIMEDEUVIMED entdecken
Crossref

RAISE: A Recommendation Framework for Service Placement in Geo-Distributed Clouds

Yifan Lin, Gongming Zhao, Hongli Xu, Jingzhou Wang, Luyao Luo, Xi Yang, Yuanpu Zhang, Wenpeng Zhu

IEEE Transactions on Networking · 2026 · 34 · 8027-8042

DOI: 10.1109/ton.2026.3740926 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Networking, 2026. Verfasst von Yifan Lin und 7 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: National Science Foundation of China, Basic Research Program of Jiangsu, Youth Innovation Promotion Association of Chinese Academy of Science.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the increasing number of enterprises establishing global data centers, the demand for efficient service placement across these geographically distributed data centers continues to rise. This demand is driven by the need to optimize various objectives, such as access delay, transmission delay, and operational cost. Existing service placement approaches typically provide a single placement plan optimizing only one objective, which fails to accommodate the diverse preferences and requirements of users. To address this gap, we introduce RAISE, a novel recommendation framework that provides a diverse set of service placement recommendations, allowing users to select a placement that best aligns with their specific preferences. However, generating such a diverse set of recommendations is computationally challenging due to the exponentially large solution space and the sparsity of valid placements. To tackle this problem, RAISE first formulates the problem as an integer non-linear program, then prunes the solution space by defining searching rules based on user requirements. Finally, it employs a subgraph-isomorphism-based algorithm to discover a diverse set of placement solutions. Extensive trace-driven simulations based on production microservice traces, real-world cloud topologies, electricity-price traces, and a small-scale deployment validation show that RAISE returns broader Pareto frontiers, higher user-satisfaction scores, and stronger deadline robustness than single-objective and multi-objective baselines.

Förderung

  • National Science Foundation of China
  • Basic Research Program of Jiangsu
  • Youth Innovation Promotion Association of Chinese Academy of Science
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Lin, Y., Zhao, G., Xu, H., Wang, J., Luo, L., Yang, X., Zhang, Y., & Zhu, W. (2026). RAISE: A Recommendation Framework for Service Placement in Geo-Distributed Clouds. IEEE Transactions on Networking, 34, 8027-8042. https://doi.org/10.1109/ton.2026.3740926

Vancouver

Lin Y, Zhao G, Xu H, Wang J, Luo L, Yang X, et al. RAISE: A Recommendation Framework for Service Placement in Geo-Distributed Clouds. IEEE Transactions on Networking. 2026;34:8027-8042. doi:10.1109/ton.2026.3740926

BibTeX
@article{Lin2026,
  author = {Lin, Yifan and Zhao, Gongming and Xu, Hongli and Wang, Jingzhou and Luo, Luyao and Yang, Xi and Zhang, Yuanpu and Zhu, Wenpeng},
  title = {RAISE: A Recommendation Framework for Service Placement in Geo-Distributed Clouds},
  journal = {IEEE Transactions on Networking},
  year = {2026},
  volume = {34},
  pages = {8027--8042},
  doi = {10.1109/ton.2026.3740926}
}
Kostenlos starten →