Cloudflare Launches Origin Range Requests for Large-File Caching

Cloudflare·23 Oct 2026·2 min read

#cloudflare

Cloudflare has introduced Origin Range Requests, a self-service feature that rebuilds how its cache handles HTTP range requests for large files. The problem was practical: a video player seeking to the middle of a movie asks for only a few megabytes, but under the old range caching implementation a single request could cause Cloudflare to pull 128 MiB from the origin and move it between data centers — sometimes after the viewer had already left. At CDN scale, that naive caching wastes both network bandwidth and disk operations.

The new implementation is now available on all plans, with no range size to tune and no support ticket required. Developers enable it through a Cache Rule, and Cloudflare manages the rest.

  • Origin Range Requests is available self-service on all plans via a Cache Rule.
  • There is no range size parameter to configure and no need to contact support.
  • For R2, Cloudflare's object storage, p99 time to first byte dropped by about half.
  • A follow-up improvement cut tiered cache traffic for R2 and similar workloads by about 6%.
  • Requests still work the same way over the wire: a normal GET carrying a Range header such as Range: bytes=1048576-2097151 returns 206 Partial Content when supported.

For developers, the change removes a tuning burden that previously required reasoning about object sizes, seek patterns and cache topology. Streaming media, large downloads and object-storage backends benefit automatically once the Cache Rule is in place, with lower latency at the tail and less origin and inter-data-center traffic. Because the fix is applied at the cache layer rather than in application code, teams serving partial content do not need to change their origin logic or client behavior — they just opt in and let Cloudflare handle the range caching.

Source: https://blog.cloudflare.com/origin-range-requests/

Related

auto-curated · source linked above← all news