Accelerating Sketch-based End-Host Traffic Measurement with Automatic DPU Offloading
Xiang Chen, Xi Sun, Wenbin Zhang, Xin Yao, Zizheng Wang, Hongyan Liu, Qun Huang, Gaoning Pan, Xuan Liu, Haifeng Zhou, Chunming Wu
摘要
Sketch-based traffic measurement is a crucial building block for monitoring traffic statistics and ensuring the quality of services of end-host applications. However, existing approaches for building sketches in end-hosts exhibit poor packet processing performance or high CPU consumption. In this paper, we propose MPU, which automatically offloads sketch-based measurement to the emerging hardware, DPU. MPU consists of a sketch analyzer that profiles sketch resource consumption and an optimization framework that formulates the offloading problem and maximizes sketch performance on DPU. We implement MPU on the NVIDIA BlueField DPU. Our testbed results indicate that MPU achieves 85% lower per-packet processing latency and 47% higher traffic measurement accuracy when compared to existing approaches.
问问这篇 Paper
问问你的智能体。
Lune 读过与它相关的顶会 Paper,每个回答都会注明依据哪几篇。
引用它的顶会 Paper1
问问它们各自怎么用它相关 Paper
- OctoSketch: Enabling Real-Time, Continuous Network Monitoring over Multiple CoresYinda Zhang, Peiqing Chen, Zaoxing LiuNSDI 2024 · 被引用 18 次
- SketchPipe: Toward Accurate Sketch-based Network Measurement on Multi-Pipeline Switches with Splitless Sketch PlacementXiang Chen, Longlong Zhu, Linying Zheng, Hongyang Du 等NSDI 2026
- Holm: A DPU-Based Robust Host-Side Latency Monitoring with Low-Overhead and Selective Full-CoverageHaifeng Zhou, Di Wang, Xi Sun, Wenbin Zhang 等INFOCOM 2026
- SketchLib: Enabling Efficient Sketch-based Monitoring on Programmable SwitchesHun Namkung, Zaoxing Liu, Daehyeok Kim, Vyas Sekar 等NSDI 2022
- Sketchovsky: Enabling Ensembles of Sketches on Programmable SwitchesHun Namkung, Zaoxing Liu, Daehyeok Kim, Vyas Sekar 等NSDI 2023 · 被引用 44 次
