HybridMesh: A Hardware-software Hybrid Approach for Accelerating Service Mesh Ingress
Myoungsung You, Jaehyun Nam, Minjae Seo, Taejune Park, Seungwon Shin
摘要
Service meshes have become essential for enabling microservice communication in cloud environments. Despite their benefits, they impose significant network overhead. We identify that the ingress gateway, the main entry point for external traffic, has emerged as a major performance bottleneck. This overhead stems from two key limitations of current ingress gateways: (1) CPU-intensive traffic analysis and (2) a complex packet forwarding path. Our analysis shows that these inefficiencies can reduce network throughput by up to 4× while substantially increasing CPU utilization. In response, we propose HybridMesh, a hardware-software hybrid ingress gateway that leverages a SmartNIC for high-performance traffic analysis and efficient traffic routing. This process is augmented by a lightweight CPU-side proxy that provides various traffic management features not suitable for Smart-NIC execution. Evaluations show that HybridMesh outperforms existing ingress gateways, achieving a 4.4× increase in HTTP throughput. Furthermore, compared to software-only and hardware-only designs, our hybrid design delivers 2.3× higher cost-effectiveness and 2.8× better power-efficiency.
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。
它引用的顶会 Paper6
- SPRIGHT: extracting the server from serverless computing! high-performance eBPF-based event-driven, shared-memory processingShixiong Qi, Leslie Monis, Ziteng Zeng, Ian-Chin Wang 等SIGCOMM 2022 · 被引用 85 次
- Rearchitecting the TCP Stack for I/O-Offloaded Content DeliveryTaehyun Kim, Deondre Martin Ng, Junzhi Gong, Youngjin Kwon 等NSDI 2023 · 被引用 42 次
- Remote Procedure Call as a Managed System ServiceJingrong Chen, Yongji Wu, Shihan Lin, Yechen Xu 等NSDI 2023 · 被引用 30 次
- LogNIC: A High-Level Performance Model for SmartNICsZerui Guo, Jiaxin Lin, Yuebin Bai, Daehyeok Kim 等MICRO 2023 · 被引用 17 次
- Conspirator: SmartNIC-Aided Control Plane for Distributed ML WorkloadsYunming Xiao, Diman Zad Tootaghaj, Aditya Dhakal, Lianjie Cao 等USENIX ATC 2024 · 被引用 13 次
相关 Paper
- Copper and Wire: Bridging Expressiveness and Performance for Service Mesh PoliciesDivyanshu Saxena, William Zhang, Shankara Pailoor, Isil Dillig 等ASPLOS 2025 · 被引用 3 次
- Enabling Fast and Stable Service Mesh Communication via Piggyback Layer-7 Traffic Control on Programmable SwitchesGonglong Chen, Jiacong Li, Yuxin Xu, Baiyan Ke 等INFOCOM 2026
- Pegasus: Transparent and Unified Kernel-Bypass Networking for Fast Local and Remote CommunicationDinglan Peng, Congyu Liu, Tapti Palit, Anjo Vahldiek-Oberwagner 等EuroSys 2025 · 被引用 6 次
- Canal Mesh: A Cloud-Scale Sidecar-Free Multi-Tenant Service Mesh ArchitectureEnge Song, Yang Song, Chengyun Lu, Tian Pan 等SIGCOMM 2024 · 被引用 30 次
- Tai Chi: A General High-Efficiency Scheduling Framework for SmartNICs in Hyperscale CloudsBang Di, Yun Xu, Kaijie Guo, Yibin Shen 等SOSP 2025
