Cloudflare launches Origin Range Requests, cutting R2 p99 TTFB by half
Cloudflare Blog · rss · 2026-10-24
Cloudflare rebuilt how its cache handles HTTP range requests and launched Origin Range Requests as a self-service feature on all plans.
The problem
- The old nginx-style fixed-slice design could pull 128 MiB from origin for a few-MB seek, and the range size was part of the cache key — changing it invalidated entries
- Small ranges caused origin-request and memory pressure; large ranges wasted bandwidth; concurrent requests in the same slice blocked each other
The new design
- Ranges align to Pingora's fixed-size disk chunks — no size tuning, no support tickets
- A coordinator composes responses from cached prefixes, in-flight downloads, and freshly fetched suffixes, validating ETag/If-Range consistency
- In-flight downloads are shared across concurrent requests for popular large files
Results
- R2 p99 time-to-first-byte dropped by about half
- Tiered cache traffic for R2-like workloads cut by a further 6%
- Enabled via a Cache Rule on any plan
More from Infra
- Baseten unveils Project Beacon to build safety tooling into the inference layer for open models — baseten · 2026-10-24
- Conflicting RSI evidence: NVIDIA says AI-assisted CUDA optimization went 5x, then 2x in a month — soumitrashukla9 · 2026-10-24
- Miles v0.1.3 cuts first RL weight update from 99.5s to 3.9s, shares GPUs across train/rollout — hsu_byron · 2026-10-24
- 98%+ Tensor Core utilization shown in LLM training run, drawing admiration — gharik · 2026-10-24
- We spent millions on AI apps with zero revenue — but our old GPUs are now worth more than we paid — stephen280up · 2026-10-24
- Polymarket puts 24% odds on AI data centers launching to space by end of next year — Polymarket · 2026-10-24