Autonomous NIC offloads
Boris Pismenny, Haggai Eran, Aviad Yehezkel, Liran Liss, Adam Morrison, Dan Tsafrir
摘要
CPUs routinely offload to NICs network-related processing tasks like packet segmentation and checksum. NIC offloads are advantageous because they free valuable CPU cycles. But their applicability is typically limited to layer≤4 protocols (TCP and lower), and they are inapplicable to layer-5 protocols (L5Ps) that are built on top of TCP. This limitation is caused by a misfeature we call "offload dependence, " which dictates that L5P offloading additionally requires offloading the underlying layer≤4 protocols and related functionality: TCP, IP, firewall, etc. The dependence of L5P offloading hinders innovation, because it implies hard-wiring the complicated, everchanging implementation of the lower-level protocols.
We propose "autonomous NIC offloads, " which eliminate offload dependence. Autonomous offloads provide a lightweight softwaredevice architecture that accelerates L5Ps without having to migrate the entire layer≤4 TCP/IP stack into the NIC. A main challenge that autonomous offloads address is coping with out-of-sequence packets. We implement autonomous offloads for two L5Ps: (i) NVMeover-TCP zero-copy and CRC computation, and (ii) https authentication, encryption, and decryption. Our autonomous offloads increase throughput by up to 3.3x, and they deliver CPU consumption and latency that are as low as 0.4x and 0.7x, respectively. Their implementation is already upstreamed in the Linux kernel, and they will be supported in the next-generation of Mellanox NICs.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了最后一个公式,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 Paper21
- 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
- The benefits of general-purpose on-NIC memoryBoris Pismenny, Liran Liss, Adam Morrison, Dan TsafrirASPLOS 2022 · 被引用 28 次
- Cornflakes: Zero-Copy Serialization for Microsecond-Scale NetworkingDeepti Raghavan, Shreya Ravi, Gina Yuan, Pratiksha Thaker 等SOSP 2023 · 被引用 18 次
- AlNiCo: SmartNIC-accelerated Contention-aware Request Scheduling for Transaction ProcessingJunru Li, Youyou Lu, Qing Wang, Jiazhen Lin 等USENIX ATC 2022 · 被引用 17 次
- Network-Offloaded Bandwidth-Optimal Broadcast and Allgather for Distributed AIMikhail Khalilov, Salvatore Di Girolamo, Marcin Chrapek, Rami Nudelman 等SC 2024 · 被引用 15 次
它引用的顶会 Paper5
- AccelTCP: Accelerating Network Applications with Stateful TCP OffloadingYoungGyoun Moon, SeungEon Lee, Muhammad Asim Jamshed, KyoungSoo ParkNSDI 2020 · 被引用 121 次
- 1RMA: Re-envisioning Remote Memory Access for Multi-tenant DatacentersArjun Singhvi, Aditya Akella, Dan Gibson, Thomas F. Wenisch 等SIGCOMM 2020 · 被引用 70 次
- TCP ≈ RDMA: CPU-efficient Remote Storage Access with i10Jaehyun Hwang, Qizhe Cai, Ao Tang, Rachit AgarwalNSDI 2020 · 被引用 70 次
- LeapIO: Efficient and Portable Virtual NVMe Storage on ARM SoCsHuaicheng Li, Mingzhe Hao, Stanko Novakovic, Vaibhav Gogte 等ASPLOS 2020 · 被引用 58 次
- IOctopus: Outsmarting Nonuniform DMAIgor Smolyar, Alex Markuze, Boris Pismenny, Haggai Eran 等ASPLOS 2020 · 被引用 16 次
相关 Paper
- How to Hardware Accelerate Your 5G CUXin Zhe Khooi, Satis Kumar Permal, Cha Hwan Song, Nishant Budhdev 等INFOCOM 2026
- F4T: A Fast and Flexible FPGA-based Full-stack TCP Acceleration FrameworkJunehyuk Boo, Yujin Chung, Eunjin Baek, Seongmin Na 等ISCA 2023 · 被引用 8 次
- SmartNS: Enabling Line-rate and Flexible Network Stack with SmartNICXuzheng Chen, Jie Zhang, Baolin Zhu, Xueying Zhu 等EuroSys 2026
- Foxhound: Server-Grade Observability for Network-Augmented ApplicationsLucas Castanheira, Alberto Schaeffer-Filho, Theophilus A. BensonEuroSys 2023 · 被引用 1 次
- Ensō: A Streaming Interface for NIC-Application CommunicationHugo Sadok, Nirav Atre, Zhipeng Zhao, Daniel S. Berger 等OSDI 2023
