Unleashing Automated Congestion Control Customization in the Wild
Amit Cohen, Lev Gloukhenki, Ravid Hadar, Eden Itah, Yehuda Shvut, Michael Schapira
TL;DR
This work tackles the problem that universal congestion control rules often underperform across diverse services and networks, leading to QoE gaps. It proposes an automated CC customization framework built on a configurable PCC Vivace, with a cloud-based engine that optimizes CC parameters using Contextual Continuum-Armed Bandits and surrogate or real QoE rewards. The approach is implemented as edge CC modules plus a centralized customization engine, with practical enhancements to speed up decision making, ensure safety, and handle inaccurate gradients. Empirical results from multi-region deployments across streaming, gaming, VoD, and connected cars demonstrate substantial QoE gains, lower rebuffering, and improved latency-throughput tradeoffs, validating the practicality and impact of automated CC customization in real-world networks.
Abstract
Congestion control (CC) crucially impacts user experience across Internet services like streaming, gaming, AR/VR, and connected cars. Traditionally, CC algorithm design seeks universal control rules that yield high performance across diverse application domains and networks. However, varying service needs and network conditions challenge this approach. We share operational experience with a system that automatically customizes congestion control logic to service needs and network conditions. We discuss design, deployment challenges, and solutions, highlighting performance benefits through case studies in streaming, gaming, connected cars, and more. Our system leverages PCC Vivace, an online-learning based congestion control protocol developed by researchers. Hence, along with insights from customizing congestion control, we also discuss lessons learned and modifications made to adapt PCC Vivace for real-world deployment.