Achieving Per-Flow Fairness and High Utilization With Limited Priority Queues in Data Center
Achieving Per-Flow Fairness and High Utilization With Limited Priority Queues in Data Center
复制标题
DOI：
10.1109/tnet.2022.3172749
复制
发表时间：
2022-10
期刊：
影响因子：
--
通讯作者：
Jingling Liu;Jiawei Huang;Zhaoyi Li;Yijun Li;Jianxin Wang;Tian He
中科院分区：
文献类型：
--
作者：
Jingling Liu;Jiawei Huang;Zhaoyi Li;Yijun Li;Jianxin Wang;Tian He
现代数据中心通常托管具有不同网络需求的多个应用。近似公平队列(AFQ)利用交换机中的多个优先级队列来近似理想的公平队列，以向数千个穿越流提供公平的带宽分配。然而，由于可编程交换机中的队列数量有限，AFQ很容易经历高丢包率和低链路利用率。在本文中，我们提出了弹性公平队列(EFQ)，它利用有限的优先级队列来灵活地实现高网络利用率和公平带宽分配。EFQ在优先级队列中为每个分组动态分配空闲缓冲区空间，以在不牺牲流级公平性的情况下获得高利用率。仿真实验和实际实现的结果表明，与现有的公平带宽分配机制相比，EFQ将平均流完成时间减少了82%。
Modern data centers often host multiple applications with diverse network demands. To provide fair bandwidth allocation to several thousand traversing flows, Approximate Fair Queueing (AFQ) utilizes multiple priority queues in switch to approximate ideal fair queueing. However, due to limited number of queues in programmable switches, AFQ easily experiences high packet loss and low link utilization. In this paper, we propose Elastic Fair Queueing (EFQ), which leverages limited priority queues to flexibly achieve both high network utilization and fair bandwidth allocation. EFQ dynamically assigns the free buffer space in priority queues for each packet to obtain high utilization without sacrificing flow-level fairness. The results of simulation experiments and real implementations show that EFQ reduces the average flow completion time by up to 82% over the state-of-the-art fair bandwidth allocation mechanisms.