The challenges of deploying Apple's Low Latency HLS In Real Life
summary
Reports using CMAF for ingest and a pipeline with Media Excel transcoders, Unified Streaming origin, nginx cache, Akamai CDN and Apple AVPlayer, implement 320 ms parts and HTTP/2 push to serve Low Latency‑HLS. End‑to‑end testing with burnt timecode and validators shows stable live latencies around three seconds, with encoder work adding ~1s. Playlist deltas, rendition reports and skip tags reduce overhead; CDN HTTP/2 push and player maturity remain the main blockers to wide rollout.
AI-written summary · about
chapters (24)
- 0:00Intro
- 0:24Outline
- 2:07Why Apple HLS
- 2:44How we fit in
- 3:07Highlights of Apple HLS
- 4:13Traditional HLS
- 5:00Encoder vs Player
- 5:30Partial Segments
- 6:06Part Tags
- 6:33Full Playlist
- 6:55Part Size
- 7:32Latency Test
- 8:12More Requests
- 8:29Blocking Reload
- 8:53Requesting Playlists
- 9:55Playlist
- 11:31Workflow
- 12:57Playlists
- 13:34Cache
- 13:48CDN
- 14:27Media Stream Validator
- 15:54Akamai Test Player
- 16:35AV Bait Error Codes
- 18:55Are we ready to ship
related talks
19mLow Latency Live from a Different Vantage Point
16mCDN Challenges of HTTP-based Low Latency Live Streaming Delivery
23m3 Second End to End Latency At Scale
44mLive Video in a Modern Web Environment
36mCHUNKY MONKEY
18mHow we accelerated Live 2 VOD delivery from 15 minutes to under a minute
12mLHLS(.js): Why Hls.js is standardizing low-latency streaming
16mImproving Video Delivery and Streaming Performance with...