L2BM: Switch Buffer Management for Hybrid Traffic in Data Center Networks
Yi Liu,Jiangping Han,Kaiping Xue,Ruidong Li,Jian Li
DOI: https://doi.org/10.1109/icdcs57875.2023.00076
2023-01-01
Abstract:With Remote Direct Memory Access (RDMA) extended to commercial Ethernet, modern Data Center Networks (DCNs) carry both traditional TCP and RDMA, to support diversified application requirements. RDMA flows are guaranteed lossless transmission through Priority-based Flow Control (PFC), while TCP flows are generally lossy traffic with packet loss. However, TCP is prone to excessively occupy the shared buffer, frequently triggering PFC pause frames and overflows at switches, damaging the performance of RDMA, which expose the vulnerability of existing buffer management policies. In this paper, we propose L2BM, a buffer management algorithm for shared-memory switches to support dynamic hybrid traffic. L2BM utilizes the average occupying time of packets in each ingress queues, to perceive the congestion states timely at ingress ports, allocating the ingress pool fairly and flexibly. Based on the perception, L2BM allocates more buffer for ingress queues with faster drain and lower congestion degrees to absorb micro-burst and reduce pause frames, less buffer for long-occupied queues to prevent excessive injection. As a result, L2BM achieves low tail latency, high burst traffic absorption capacity and low buffer occupancy. Evaluations show that L2BM enable to cut the tail latency of RDMA traffic by 50% at high workloads, reduce the buffer occupancy by 40% and decrease average query delay by 57%, while ensuring few PFC pause frames and maintaining good performance of TCP flows.