EUVIMEDEUVIMED entdecken
Crossref

R AIN : RDMA-assisted In-Network Scheduling for Microsecond-scale Workloads

Zhihuang Ma, Xingming Cui, Xiaoliang Chen, Zuqing Zhu

Proceedings of the ACM on Networking · 01.06.2026 · 4 (CoNEXT2) · 1-21

DOI: 10.1145/3808670 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the ACM on Networking, online erschienen am 01.06.2026. Verfasst von Zhihuang Ma und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: The National Key R&D Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern data center applications increasingly require microsecond-scale service time with strict tail latency requirements, which can hardly be realized with existing in-network task schedulers due to their inherent limitations. Specifically, software-based schedulers struggle to balance throughput and latency, while switch-based designs either lack global coordination, rely on packet recirculation heavily, or only offer limited support for large tasks. In light of these restrictions of the state-of-the-arts (SOTAs), we, in this work, propose Rain, an RDMA-assisted in-network scheduler built atop programmable switches that maintains centralized queues while bounding worker-local queues. Rain introduces a bidirectional on-switch queuing mechanism to buffer and match tasks and worker-issued tokens directly in the switch, avoiding worker-side polling and approximating the optimal behavior of join-bounded-shortest-queue without global aggregation. A switch-driven RDMA engine pre-writes arbitrarily large tasks via one-sided WRITE multicasts, keeping only compact metadata on the switch. Slice-aware scheduling further localizes decisions to more homogeneous queues, reducing dispersion-induced head-of-line blocking. Moreover, our study reveals that real-world systems can diverge from theoretical predictions: shallower worker queues do not always improve tail latency. Leveraging this insight, Rain incorporates an adaptive scheduling strategy to optimize worker queue depths and worker-to-slice mappings at runtime. Evaluations with the real-world application RocksDB show that Rain achieves 1.75× higher throughput than the best-performing SOTA while satisfying the same tail latency requirement.

Förderung

  • The National Key R&D Program of China

Zitieren

APA

Ma, Z., Cui, X., Chen, X., & Zhu, Z. (2026). R AIN : RDMA-assisted In-Network Scheduling for Microsecond-scale Workloads. Proceedings of the ACM on Networking, 4(CoNEXT2), 1-21. https://doi.org/10.1145/3808670

Vancouver

Ma Z, Cui X, Chen X, Zhu Z. R AIN : RDMA-assisted In-Network Scheduling for Microsecond-scale Workloads. Proceedings of the ACM on Networking. 2026;4(CoNEXT2):1-21. doi:10.1145/3808670

BibTeX
@article{Ma2026,
  author = {Ma, Zhihuang and Cui, Xingming and Chen, Xiaoliang and Zhu, Zuqing},
  title = {R AIN : RDMA-assisted In-Network Scheduling for Microsecond-scale Workloads},
  journal = {Proceedings of the ACM on Networking},
  year = {2026},
  volume = {4},
  number = {CoNEXT2},
  pages = {1--21},
  doi = {10.1145/3808670}
}
Kostenlos starten →