Loki
Authors:
Open Access
- 25 October 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 775-788
Abstract
Maximizing the quality of experience (QoE) for real-time video is a long-standing challenge. Traditional video transport protocols, represented by a few deterministic rules, can hardly adapt to the heterogeneous and highly dynamic modern Internet. Emerging learning-based algorithms have demonstrated potential to meet the challenge. However, our measurement study reveals an alarming long tail performance issue: these algorithms tend to be bottle-necked by occasional catastrophic events due to the built-in exploration mechanisms. In this work, we propose Loki, which improves the robustness of learning-based model by coherently integrating it with a rule-based algorithm. To enable integration at feature level, we first reverse-engineer the rule-based algorithm into an equivalent "black-box" neural network. Then, we design a dual-attention feature fusion mechanism to fuse it with a reinforcement learning model. We train Loki in a commercial real-time video system through online learning, and evaluate it over 101 million video sessions, in comparison to state-of-the-art rule-based and learning-based solutions. The results show that Loki improves not only the average but also the tail performance substantially (26.30% to 44.24% reduction of stall rate and 1.76% to 2.17% increase in video throughput at 95-percentile).Keywords
Funding Information
- Youth Top Talent Support Program
- AIR Program
- 111 Project (B18008)
- NSFC (61772084, 61720106007, 61832010)
- Innovation Research Group Project of NSFC (61921003)
- BUPT Excellent Ph.D. Students Foundation (CX2020204)
This publication has 23 references indexed in Scilit:
- Deep Learning towards Mobile Applications2018
- Neural Adaptive Video Streaming with Pensieve2017
- Congestion Control for Web Real-Time CommunicationIEEE/ACM Transactions on Networking, 2017
- BBRCommunications of the ACM, 2017
- Adaptive Congestion Control for Unpredictable Cellular Networks2015
- Bufferbloat systematic analysis2014
- TCP ex machina2013
- PROTEUS2013
- An experimental evaluation of rate-adaptive video players over HTTPSignal Processing: Image Communication, 2012
- CUBICACM SIGOPS Operating Systems Review, 2008