Towards efficient dynamic Ad insertion in the Edge
summary
Claims a working integration of HLS low‑latency playlist manipulation with a CDN that keeps players at the edge of the stream while handling ad server communication and timing constraints. Uses pre-generated chunked ads and per‑ad playlists with the last three segments and explicit chunk entries so the DI Gateway located next to the edge cache can splice in ads without added latency. Tests show HLS stricter than DASH, uses tens or hundreds ad pools and short slates (tens of ms to ~2s) to fill gaps.
AI-written summary · about
chapters (10)
- 0:00Dynamic ad insertion intro
- 1:51Preparing ad markers
- 3:58Low-latency challenges & solutions
- 5:44Low-latency architecture
- 8:38Ad server playlist prep
- 10:22Edge gateway insertion
- 12:28Real-time ad insertion handling
- 14:39Slates for missing durations
- 16:01Thematic ad pools customization
- 19:15Conclusion & future work
related talks
15mPseudo-Interstitials: Playback flexibility for legacy devices
19mLow Latency Live from a Different Vantage Point
24mDynamic content insertion into live streams across the 3 main streaming formats
10mEnhancing CDN Performance and Cutting Egress Costs in Large Video Libraries
16mImproving Video Delivery and Streaming Performance with...
45mHigh Performance Ads - Is It
19mAnother talk about ads, but this time what if we talked...
12mLHLS(.js): Why Hls.js is standardizing low-latency streaming