EUVIMEDEUVIMED entdecken
Crossref

Reinforcement Learning for Datacenter Congestion Control

Chen Tessler, Yuval Shpigelman, Gal Dalal, Amit Mandelbaum, Doron Haritan Kazakov, Benjamin Fuhrer, Gal Chechik, Shie Mannor

ACM SIGMETRICS Performance Evaluation Review · 17.01.2022 · 49 (2) · 43-46

DOI: 10.1145/3512798.3512815 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM SIGMETRICS Performance Evaluation Review, online erschienen am 20.01.2022. Verfasst von Chen Tessler und 7 weiteren Autor:innen. Die Arbeit wurde bisher 34-mal zitiert (Crossref) und nennt 24 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We approach the task of network congestion control in datacenters using Reinforcement Learning (RL). Successful congestion control algorithms can dramatically improve latency and overall network throughput. Until today, no such learning-based algorithms have shown practical potential in this domain. Evidently, the most popular recent deployments rely on rule-based heuristics that are tested on a predetermined set of benchmarks. Consequently, these heuristics do not generalize well to newly-seen scenarios. Contrarily, we devise an RL-based algorithm with the aim of generalizing to different configurations of real-world datacenter networks. We overcome challenges such as partial-observability, nonstationarity, and multi-objectiveness. We further propose a policy gradient algorithm that leverages the analytical structure of the reward function to approximate its derivative and improve stability. We show that this scheme outperforms alternative popular RL approaches, and generalizes to scenarios that were not seen during training. Our experiments, conducted on a realistic simulator that emulates communication networks' behavior, exhibit improved performance concurrently on the multiple considered metrics compared to the popular algorithms deployed today in real datacenters. Our algorithm is being productized to replace heuristics in some of the largest datacenters in the world.

Zitieren

APA

Tessler, C., Shpigelman, Y., Dalal, G., Mandelbaum, A., Kazakov, D. H., Fuhrer, B., Chechik, G., & Mannor, S. (2022). Reinforcement Learning for Datacenter Congestion Control. ACM SIGMETRICS Performance Evaluation Review, 49(2), 43-46. https://doi.org/10.1145/3512798.3512815

Vancouver

Tessler C, Shpigelman Y, Dalal G, Mandelbaum A, Kazakov DH, Fuhrer B, et al. Reinforcement Learning for Datacenter Congestion Control. ACM SIGMETRICS Performance Evaluation Review. 2022;49(2):43-46. doi:10.1145/3512798.3512815

BibTeX
@article{Tessler2022,
  author = {Tessler, Chen and Shpigelman, Yuval and Dalal, Gal and Mandelbaum, Amit and Kazakov, Doron Haritan and Fuhrer, Benjamin and Chechik, Gal and Mannor, Shie},
  title = {Reinforcement Learning for Datacenter Congestion Control},
  journal = {ACM SIGMETRICS Performance Evaluation Review},
  year = {2022},
  volume = {49},
  number = {2},
  pages = {43--46},
  doi = {10.1145/3512798.3512815}
}
Kostenlos starten →