Cache-hit ratio is the percentage of viewer requests answered directly from the CDN edge cache rather than fetched from the origin. A high ratio (often 95%+ for popular VOD) means the origin barely sees traffic, keeping both latency and expensive origin egress low; a low ratio means the origin is doing work the CDN should absorb.
It is driven by cache-key design (identical requests must map to the same object), content popularity, and TTL tuning - and easily wrecked by personalization or per-user tokens that fragment objects. Because delivery cost scales with origin pulls, improving cache-hit ratio is one of the cheapest, highest-impact levers in OTT cost engineering.
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
Cache-hit ratio is the dial that sets most of your delivery cost, because every miss goes back to the origin. We wrote seven volumes on the video stack; for cache-hit ratio start with Volume 3 on CDNs, caching, and the economics of delivery, then Volume 2 on segment packaging, with rendition count from encoding in Volume 1. Written by Nikolay Sapunov, CEO at Fora Soft.
Origin bills climbing because your CDN cache-hit ratio is too low?
We will review your cache keys, tokens, and segment layout and send changes with the expected hit-ratio gain. 250+ video products shipped since 2005.
