JitBright: towards Low-Latency Mobile Cloud Rendering through Jitter Buffer Optimization

Yuankang Zhao, Qinghua Wu, Gerui Lv, Furong Yang, Jiuhai Zhang, Feng Peng, Yanmei Liu, Zhenyu Li, Ying Chen, Hongyu Guo, Gaogang Xie

Abstract: Low-latency cloud rendering services use high-performance servers to provide mobile device users with exquisite graphics and convenient access experiences. Due to the complexity of the system and the diversity of impacting factors, identifying system bottlenecks has become a significant challenge. To demystify system performance, we build an online cloud rendering system to measure the latency distribution of its key components.Our real-world measurement study reveals that the primary factor causing increased motion-to-photon (MTP) latency is the receive-to-composition (R2C) latency at the client, which is primarily caused by the ineffective jitter buffer management strategy. Based on these findings, we propose JitBright, a systematic jitter buffer optimization that deflates MTP latency. JitBright incorporates adaptive gain and proactive keyframe requests. Large-scale A/B tests involving over 12,000 users demonstrate that JitBright successfully reduces MTP latency while improving playout smoothness. Specifically, JitBright increases the proportion of sessions that meet MTP latency requirements by 6%-27%.