Revisiting Network Support for RDMA

TL;DR

This paper challenges the necessity of Priority Flow Control for RDMA over Ethernet, proposing an improved NIC design that eliminates PFC and enhances performance with minimal resource overhead.

Contribution

It introduces IRN, a new NIC design that removes PFC requirements for RDMA, demonstrating performance gains and low implementation costs.

Findings

IRN outperforms RoCE with PFC by 6-83% in typical scenarios.

IRN eliminates PFC, reducing network issues like congestion spreading.

Implementation overhead of IRN is about 3-10% of NIC resources.

Abstract

The advent of RoCE (RDMA over Converged Ethernet) has led to a significant increase in the use of RDMA in datacenter networks. To achieve good performance, RoCE requires a lossless network which is in turn achieved by enabling Priority Flow Control (PFC) within the network. However, PFC brings with it a host of problems such as head-of-the-line blocking, congestion spreading, and occasional deadlocks. Rather than seek to fix these issues, we instead ask: is PFC fundamentally required to support RDMA over Ethernet? We show that the need for PFC is an artifact of current RoCE NIC designs rather than a fundamental requirement. We propose an improved RoCE NIC (IRN) design that makes a few simple changes to the RoCE NIC for better handling of packet losses. We show that IRN (without PFC) outperforms RoCE (with PFC) by 6-83% for typical network scenarios. Thus not only does IRN eliminate the need for PFC, it improves performance in the process! We further show that the changes that IRN introduces can be implemented with modest overheads of about 3-10% to NIC resources. Based on our results, we argue that research and industry should rethink the current trajectory of network support for RDMA.