EUVIMEDEUVIMED entdecken
Crossref

Improving Storage Systems Using Machine Learning

Ibrahim Umit Akgun, Ali Selman Aydin, Andrew Burford, Michael McNeill, Michael Arkhangelskiy, Erez Zadok

ACM Transactions on Storage · 28.02.2023 · 19 (1) · 1-30

DOI: 10.1145/3568429 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Storage, online erschienen am 19.01.2023. Verfasst von Ibrahim Umit Akgun und 5 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 106 Literaturangaben. Gefördert von: NSF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Operating systems include many heuristic algorithms designed to improve overall storage performance and throughput. Because such heuristics cannot work well for all conditions and workloads, system designers resorted to exposing numerous tunable parameters to users—thus burdening users with continually optimizing their own storage systems and applications. Storage systems are usually responsible for most latency in I/O-heavy applications, so even a small latency improvement can be significant. Machine learning (ML) techniques promise to learn patterns, generalize from them, and enable optimal solutions that adapt to changing workloads. We propose that ML solutions become a first-class component in OSs and replace manual heuristics to optimize storage systems dynamically. In this article, we describe our proposed ML architecture, called KML. We developed a prototype KML architecture and applied it to two case studies: optimizing readahead and NFS read-size values. Our experiments show that KML consumes less than 4 KB of dynamic kernel memory, has a CPU overhead smaller than 0.2%, and yet can learn patterns and improve I/O throughput by as much as 2.3× and 15× for two case studies—even for complex, never-seen-before, concurrently running mixed workloads on different storage devices.

Förderung

  • NSF

Zitieren

APA

Akgun, I. U., Aydin, A. S., Burford, A., McNeill, M., Arkhangelskiy, M., & Zadok, E. (2023). Improving Storage Systems Using Machine Learning. ACM Transactions on Storage, 19(1), 1-30. https://doi.org/10.1145/3568429

Vancouver

Akgun IU, Aydin AS, Burford A, McNeill M, Arkhangelskiy M, Zadok E. Improving Storage Systems Using Machine Learning. ACM Transactions on Storage. 2023;19(1):1-30. doi:10.1145/3568429

BibTeX
@article{Akgun2023,
  author = {Akgun, Ibrahim Umit and Aydin, Ali Selman and Burford, Andrew and McNeill, Michael and Arkhangelskiy, Michael and Zadok, Erez},
  title = {Improving Storage Systems Using Machine Learning},
  journal = {ACM Transactions on Storage},
  year = {2023},
  volume = {19},
  number = {1},
  pages = {1--30},
  doi = {10.1145/3568429}
}
Kostenlos starten →