Implementing HLS/DASH Content Steering at Scale
summary
Claim: content steering works at mass scale by pushing trivial decision logic to stateless edge functions while doing heavy analytics in the cloud. Implementation uses standard HLS/DASH tags, DASH.js and HLS.js players, and steering servers deployed on multiple edge platforms; a demo across three edges shows automatic switches when reported throughput falls below threshold. The design lowers operating cost, lets TTL drop from the nominal ~300s to react faster, and the open source submission to the SVTA alliance is available.
AI-written summary · about
chapters (8)
- 0:00Multi-CDN video delivery
- 1:05Content steering standard benefits
- 2:51Cloud steering challenges
- 4:00Edge steering servers
- 5:26Demo setup and features
- 7:53Demo walkthrough CDN switch
- 9:19Bandwidth throttling QoE
- 10:22Conclusion future apps
related talks
26mContent Steering with MPEG DASH
21mWhat's new in HLS
24mDynamic content insertion into live streams across the 3 main streaming formats
23mYou Deserve Low-Latency Sports and We Serve It via Content-Aware Playback
08mHow to play Dash on your HLS Player
12mLHLS(.js): Why Hls.js is standardizing low-latency streaming
37mThree Roads to Jerusalem
20mDecentralizing Video Streaming