7 January 2025 in IEEE Transactions on Networking
RDMA has been widely deployed in production datacenters. The conventional wisdom believes that the intra-host network delivers stable and high performance. However, intra-host resources witness a relative stagnation in technology trends compared to the evolving RDMA NIC (RNIC). Thus, the RNIC traffic may not get sufficient intra-host resources when it contends with CPU-to-memory traffic. A line of recent works from large-scale production datacenter operators demonstrates the emergence of intra-host congestion and associated performance collapse, which forces us to revisit the practice of intra-host congestion control. However, the ability to efficiently control RDMA intra-host networks is far less mature than inter-host networks, which brings challenges in congestion monitoring, intra-host resource allocation and RNIC traffic adjustment. In this paper, we propose RDMA intra-Host Congestion Control (RHCC), which combines CPU-to-memory traffic congestion avoidance with sub-RTT granularity and proactive RNIC traffic adjustment. RHCC ensures fast congestion avoidance and can work with different inter-host congestion control methods. We implement RHCC on commodity servers and RNICs and conduct experiments to evaluate the performance. The results show that RHCC can increase/decrease the network throughput/latency by up to 2 $\times$ and 1.4 $\times$ , respectively.
16 July 2025 in IEEE Network
Data center networks are critical infrastructures currently serving fields such as cloud computing and artificial intelligence, where congestion control algorithms play a critical role in improving network performance. In this paper, we focus on the current research status and prospects of congestion control approach in Remote Direct Memory Access (RDMA) networks and propose a new algorithm to address existing shortcomings. Firstly, we classify and summarize current congestion control algorithms, exploring the advantages and disadvantages of the end-to-end and end-network collaboration approaches. We argue that, considering the trade-off between network performance and hardware costs, attention should still be directed towards end-to-end congestion control algorithms. Then, we analyzed issues with mainstream solutions and introduced a new fast convergence congestion control algorithm, FCC. This algorithm is based on Explicit Congestion Notification (ECN) and Round-Trip Time (RTT) signals, leveraging the gradient of RTT and a Sigmoid curve to enhance response speeds. Finally, we conducted a performance evaluation of this algorithm and compared it with existing ones. Experimental results demonstrate that the FCC algorithm achieves superior performance in terms of convergence speed, fairness, and latency for short flows.
1 January 2026 in IEEE Transactions on Network and Service Management