NetLock: Fast, Centralized Lock Management Using Programmable Switches
Zhuolong Yu, Yiwen Zhang, Vladimir Braverman, Mosharaf Chowdhury, Xin Jin
摘要
Lock managers are widely used by distributed systems. Traditional centralized lock managers can easily support policies between multiple users using global knowledge, but they suffer from low performance. In contrast, emerging decentralized approaches are faster but cannot provide flexible policy support. Furthermore, performance in both cases is limited by the server capability.
We present NetLock, a new centralized lock manager that codesigns servers and network switches to achieve high performance without sacrificing flexibility in policy support. The key idea of NetLock is to exploit the capability of emerging programmable switches to directly process lock requests in the switch data plane. Due to the limited switch memory, we design a memory management mechanism to seamlessly integrate the switch and server memory. To realize the locking functionality in the switch, we design a custom data plane module that efficiently pools multiple register arrays together to maximize memory utilization We have implemented a NetLock prototype with a Barefoot Tofino switch and a cluster of commodity servers. Evaluation results show that NetLock improves the throughput by 14.0ś18.4×, and reduces the average and 99% latency by 4.7ś20.3× and 10.4ś18.7× over DSLR, a state-of-the-art RDMA-based solution, while providing flexible policy support.
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 Paper32
- CocoSketch: high-performance sketch-based measurement over arbitrary partial key queryYinda Zhang, Zaoxing Liu, Ruixin Wang, Tong Yang 等SIGCOMM 2021 · 被引用 146 次
- Programmable packet scheduling with a single queueZhuolong Yu, Chuheng Hu, Jingfeng Wu, Xiao Sun 等SIGCOMM 2021 · 被引用 100 次
- Sherman: A Write-Optimized Distributed B+Tree Index on Disaggregated MemoryQing Wang, Youyou Lu, Jiwu ShuSIGMOD 2022 · 被引用 99 次
- Electrode: Accelerating Distributed Protocols with eBPFYang Zhou, Zezhou Wang, Sowmya Dharanipragada, Minlan YuNSDI 2023 · 被引用 78 次
- Concordia: Distributed Shared Memory with In-Network Cache CoherenceQing Wang, Youyou Lu, Erci Xu, Junru Li 等FAST 2021 · 被引用 74 次
它引用的顶会 Paper2
相关 Paper
- ShiftLock: Mitigate One-sided RDMA Lock Contention via HandoverJian Gao, Qing Wang, Jiwu ShuFAST 2025 · 被引用 10 次
- Fast and Scalable In-network Lock Management Using Lock FissionHanze Zhang, Ke Cheng, Rong Chen, Haibo ChenOSDI 2024 · 被引用 9 次
- Citron: Distributed Range Lock Management with One-sided RDMAJian Gao, Youyou Lu, Minhui Xie, Qing Wang 等FAST 2023 · 被引用 13 次
- Scalable RDMA-accelerated Distributed Locks with Shared Stream AbstractionMiao Cai, Junru Shen, Xiaojian Liao, Rong Gu 等EuroSys 2026
- FARLock: Asymmetric RDMA Locking Made FairYuehao Hu, Jiatang Zhou, Tianzheng Wang, Keval VoraOSDI 2026
