RecoNIC 入门：基于 SmartNIC 的 RDMA 计算卸载
A Primer on RecoNIC: RDMA-enabled Compute Offloading on SmartNIC
AI总结：
本文介绍开源 FPGA SmartNIC 平台 RecoNIC，通过共享 RDMA 卸载引擎和可编程计算块降低数据复制开销，支持高吞吐、低延迟的近网络计算。
摘要：
如今的数据中心由成千上万台联网主机组成，每台主机都配备 CPU 以及 GPU、FPGA 等加速器。这些主机还包含以 100Gb/s 或更高速率运行的网卡（NIC），用于彼此通信。我们提出 RecoNIC，这是一个基于 FPGA、支持 RDMA 的 SmartNIC 平台，旨在通过让网络数据尽可能靠近计算，在实现计算加速的同时，最大限度减少以 CPU 为中心的加速器系统中与数据复制相关的开销。由于 RDMA 已成为改善数据中心工作负载通信的事实传输层协议，RecoNIC 集成了 RDMA 卸载引擎，以实现高吞吐量和低延迟的数据传输。开发者可以灵活地在 RecoNIC 的可编程计算块中使用 RTL、HLS 或 Vitis Networking P4 设计其加速器。这些计算块既可以访问主机内存，也可以通过 RDMA 卸载引擎访问远程对等节点的内存。此外，RDMA 卸载引擎由主机和计算块共享，这使 RecoNIC 成为一个非常灵活的平台。最后，我们已面向研究社区开源 RecoNIC，以支持基于 RDMA 的应用和用例实验。
英文摘要：
Today's data centers consist of thousands of network-connected hosts, each with CPUs and accelerators such as GPUs and FPGAs. These hosts also contain network interface cards (NICs), operating at speeds of 100Gb/s or higher, that are used to communicate with each other. We propose RecoNIC, an FPGA-based RDMA-enabled SmartNIC platform that is designed for compute acceleration while minimizing the overhead associated with data copies (in CPU-centric accelerator systems) by bringing network data as close to computation as possible. Since RDMA is the defacto transport-layer protocol for improved communication in data center workloads, RecoNIC includes an RDMA offload engine for high throughput and low latency data transfers. Developers have the flexibility to design their accelerators using RTL, HLS or Vitis Networking P4 within the RecoNIC's programmable compute blocks. These compute blocks can access host memory as well as memory in remote peers through the RDMA offload engine. Furthermore, the RDMA offload engine is shared by both the host and compute blocks, which makes RecoNIC a very flexible platform. Lastly, we have open-sourced RecoNIC for the research community to enable experimentation with RDMA-based applications and use-cases.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
