RLive: Robust Delivery System for Scaling Live Streaming Services
Yu Tian, Gerui Lv, Qinghua Wu, Ruili Fang, Yajie Peng, Zhichen Xue, Rui Han, Chuanqing Lin, Xiaofei Pang, Ri Lu, Zhenyu Li
Abstract
As the demand for streaming services surges, content delivery network (CDN) operators face increasing pressure to scale live video delivery without proportionally increasing infrastructure costs. While best-effort edge resources offer a cost-effective extension to traditional CDN capacity, their limited bandwidth and unstable performance pose significant challenges. Our operational experience shows that naively layering such resources onto existing CDN infrastructure falls short in meeting performance and scalability demands. This paper presents RLive, a robust delivery system that scales CDN capacity by integrating best-effort edge resources. RLive features a redundancy-free multi-source data plane to support reliable and cost-efficient live streaming, along with a multi-layer collaborative control plane that combines the global view with local adaptability for scalable user-to-node mapping. Deployed in ByteDance CDN to support large-scale live streaming services with hundreds of millions of daily viewers, RLive has tripled delivery capacity while reducing rebuffering events by 14.9–20.1%.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 18f36b10-fa5d-4c35-ade4-a8c6b07620eeBuilds on25
- Neural-Enhanced Live Streaming: Improving Live Video Ingest via Online LearningJaehong Kim, Youngmok Jung, Hyunho Yeo, Juncheol Ye et al.SIGCOMM 2020 · 132 citations
- XLINK: QoE-driven multi-path QUIC transport in large-scale video servicesZhilong Zheng, Yunfei Ma, Yanmei Liu, Furong Yang et al.SIGCOMM 2021 · 120 citations
- Contracting Wide-area Network Topologies to Solve Flow Problems QuicklyFiras Abuzaid, Srikanth Kandula, Behnaz Arzani, Ishai Menache et al.NSDI 2021 · 101 citations
- Teal: Learning-Accelerated Optimization of WAN Traffic EngineeringZhiying Xu, Francis Y. Yan, Rachee Singh, Justin T. Chiu et al.SIGCOMM 2023 · 95 citations
- DOTE: Rethinking (Predictive) WAN Traffic EngineeringYarin Perry, Felipe Vieira Frujeri, Chaim Hoch, Srikanth Kandula et al.NSDI 2023 · 81 citations
Related papers
- HCDN: Coordinated Stream Scheduling for Cost-Effective Live Video DeliveryLiying Wang, Jing Liu, Yuhan Zhou, Chengke Wang et al.NSDI 2026
- HyperEdge: An Edge CDN Infrastructure for Cost Efficient Video StreamingDehui Wei, Jiao Zhang, Haozhe Li, Rui Han et al.NSDI 2026 · 1 citation
- Medley: Optimizing Midgress Bandwidth for Commercial Live Streaming CDNsHaiping Wang, Wanxin Shi, Sandesh Dhawaskar Sathyanarayana, Shu Shi et al.NSDI 2026
- Enhancing Resource Management of the World's Largest PCDN System for On-Demand Video StreamingRuixiao Zhang, Haiping Wang, Shu Shi, Xiaofei Pang et al.USENIX ATC 2024 · 25 citations
- LiveNet: a low-latency video transport network for large-scale live streamingJinyang Li, Zhenyu Li, Ri Lu, Kai Xiao et al.SIGCOMM 2022 · 59 citations
