PRED: Performance-oriented Random Early Detection for Consistently Stable Performance in Datacenters

For decades, Random Early Detection (RED) has been integrated into datacenter switches as a fundamental Active Queue Management (AQM). Accurate configuration of RED parameters is crucial to achieving high throughput and low latency. However, due to the highly dynamic nature of workloads in datacenter networks, maintaining consistently high performance with statically configured RED thresholds poses a challenge. Prior art applies reinforcement learning to predict proper thresholds, but their real-world deployment has been hindered by poor tail performance caused by instability. In this paper, we propose PRED, a novel system that enables automatic and stable RED parameter adjustment in response to traffic dynamics. PRED uses two loosely coupled systems, Flow Concurrent Stabilizer (FCS) and Queue Length Adjuster (QLA), to overcome the challenges of dynamically setting RED parameters to adapt to the ever-changing traffic pattern. We perform extensive evaluations on our physical testbed and large-scale simulations. The results demonstrate that PRED can keep up with the real-time network dynamics generated by realistic workloads. For instance, compared with the static-threshold-based methods, PRED keeps 66% lower switch queue length and obtains up to 80% lower Flow Completion Time (FCT). Compared with the state-of-the-art learning-based method, PRED reduces the tail FCT by 34%.

(e.g., HPCC [7] , Swift [8] , BFC [9] )

Case-based (e.g., ECNsharp [15] , BCC [16] )

New AQM(e.g., TCD [14] ) Hard to deploy, needs new switch design Adjust RED Parameter Widely used, commercial verification Prediction model (e.g., ACC [17] ) Unstable, poor tail performance