Origin Range Requests: Serving large files without range-request tuning
Blog post from Cloudflare
Cloudflare has introduced Origin Range Requests, a self-service Cache Rules feature that improves delivery of partial requests for large files such as videos, downloads, software packages, disk images, and AI model artifacts. The feature replaces an older fixed-range caching system in which customer-configured range sizes could either generate many small origin requests or cause Cloudflare to retrieve and transfer far more data than a client needed, sometimes across multiple data centers. Built on Cloudflare’s Pingora-based cache, the new approach stores files in fixed-size disk chunks, serves already cached bytes directly, fetches only missing aligned chunks from origins, and coordinates cached, in-progress, and newly fetched data into accurate HTTP range responses. It also validates asset consistency through range, size, and entity-tag checks to avoid combining bytes from different file versions. After migration, R2 object storage saw roughly a 50% reduction in p99 time to first byte, while an additional optimization reduced tiered-cache traffic for R2 and similar workloads by about 6%. Available on all plans, the feature requires cacheable content and origin support for byte ranges, and can be enabled for selected traffic without configuring a range size.
| Trend | Post Mentions | Total Month Mentions | Posts | Companies | MoM |
|---|---|---|---|---|---|
| Real-time | 2 | 2,798 | 806 | 212 | -42% |
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.