Aeolus: A Building Block for Proactive Transport in Datacenters
Shuihai Hu, Wei Bai, Gaoxiong Zeng, Zilong Wang, Baochen Qiao, Kai Chen, Kun Tan, Yi Wang
摘要
As datacenter network bandwidth keeps growing, proactive transport becomes attractive, where bandwidth is proactively allocated as "credits" to senders who then can send "scheduled packets" at a right rate to ensure high link utilization, low latency, and zero packet loss. While promising, a fundamental challenge is that proactive transport requires at least one-RTT for credits to be computed and delivered. In this paper, we show such one-RTT "pre-credit" phase could carry a substantial amount of flows at high link-speeds, but none of existing proactive solutions treats it appropriately. We present Aeolus, a solution focusing on "pre-credit" packet transmission as a building block for proactive transports. Aeolus contains unconventional design principles such as scheduled-packet-first (SPF) that de-prioritizes the first-RTT packets, instead of prioritizing them as prior work. It further exploits the preserved, deterministic nature of proactive transport as a means to recover lost first-RTT packets efficiently. We have integrated Aeolus into ExpressPass[14], NDP[18] and Homa[29], and shown, through both implementation and simulations, that the Aeolus-enhanced solutions deliver signiicant performance or deployability advantages. For example, it improves the average FCT of ExpressPass by 56%, cuts the tail FCT of Homa by 20x, while achieving similar performance as NDP without switch modifications.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper29
- Empowering Azure Storage with RDMAWei Bai, Shanim Sainul Abdeen, Ankit Agrawal, Krishan Kumar Attre 等NSDI 2023 · 被引用 117 次
- PowerTCP: Pushing the Performance Limits of Datacenter NetworksVamsi Addanki, Oliver Michel, Stefan SchmidNSDI 2022 · 被引用 116 次
- Programmable packet scheduling with a single queueZhuolong Yu, Chuheng Hu, Jingfeng Wu, Xiao Sun 等SIGCOMM 2021 · 被引用 100 次
- Congestion detection in lossless networksYiran Zhang, Yifan Liu, Qingkai Meng, Fengyuan RenSIGCOMM 2021 · 被引用 62 次
- Towards Domain-Specific Network Transport for Distributed DNN TrainingHao Wang, Han Tian, Jingrong Chen, Xinchen Wan 等NSDI 2024 · 被引用 54 次
相关 Paper
- FlexPass: A Case for Flexible Credit-based Transport for Datacenter NetworksHwijoon Lim, Jaehong Kim, Inho Cho, Keon Jang 等EuroSys 2023 · 被引用 12 次
- PPT: A Pragmatic Transport for DatacentersLide Suo, Yiren Pang, Wenxin Li, Renjie Pei 等SIGCOMM 2024 · 被引用 7 次
- Unlocking Superior Performance in Reconfigurable Data Center Networks with Credit-Based TransportFederico De Marchi, Jialong Li, Ying Zhang, Wei Bai 等SIGCOMM 2025 · 被引用 2 次
- Towards timeout-less transport in commodity datacenter networksHwijoon Lim, Wei Bai, Yibo Zhu, Youngmok Jung 等EuroSys 2021 · 被引用 28 次
- eTran: Extensible Kernel Transport with eBPFZhongjie Chen, Qingkai Meng, ChonLam Lao, Yifan Liu 等NSDI 2025 · 被引用 17 次
