Enhancing CDN Performance and Cutting Egress Costs in Large Video Libraries
summary
Claims that global content jumps to 147 zettabytes in 2024 and that operators still target 98%+ cache hit rates drive concrete work: hierarchical tiered caches with URL-based consistent hashing, Bloom-filter-based cache-state sharing, and simple counters to eliminate one-hit wonders. Edge tricks include prefetching ordered segments, transmuxing at the edge to store a single format, and experiments replacing HLS/DASH on the backbone with SRT or media-over-QUIC to cut requests lower latency and largely eliminate live cache misses.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 0:42CDN caching challenges
- 2:30How CDN caching works
- 5:21Video CDN optimizations
- 9:30Closing
related talks
16mImproving Video Delivery and Streaming Performance with...
16mCDN Challenges of HTTP-based Low Latency Live Streaming Delivery
23mContent delivery sucks, and what you can do about it
53mCDNs 101: An Introduction to Content Delivery Networks
20mDecentralizing Video Streaming
19mLow Latency Live from a Different Vantage Point
05mWeb Performance, Why it Really Matters
15mUsing The Power of Edge Compute To Enhance Streaming Delivery