About

Static Edge Networks

We run a small, deliberately unglamorous edge cache. Objects are stored close to the people requesting them, origins are asked for content as rarely as correctness allows, and everything the service does is visible in a response header.

01What the service does

Static Edge accepts requests for a customer zone at the nearest edge node, serves any fresh object it already holds, and otherwise fetches from the customer's origin and stores the result according to the caching headers on that response. That is the whole product. There is no image transformation pipeline, no edge compute runtime, and no traffic scripting language to learn.

The design goals, in order:

1. Correctness first. Standard HTTP caching semantics are honoured exactly. If the origin says a response is private, it is never stored.
2. Observable behaviour. Every response carries the cache state, the node that served it and the key that was used, so a hit ratio problem can be diagnosed with curl rather than a support ticket.
3. Fast invalidation. Purge by URL, prefix or surrogate key, propagated to every node within seconds, so short TTLs are not the only tool available.

02How it is operated

Nodes run in a handful of European and North American facilities on hardware we lease directly. Configuration changes ship through the same reviewed pipeline as software changes, and each node is drainable — traffic moves away before maintenance, not during it. Cached objects live in memory and on local disk; nothing is written to shared storage.

• Founded2021
• TeamEleven people, distributed
• LocationsFrankfurt, Amsterdam, London, Warsaw, Montreal, Dallas
• Edge software3.4, released 4 August 2026
• Uptime target99.95% monthly, measured at the edge
• Supportsupport at static-edge-cdn.xyz
• Security reportssecurity at static-edge-cdn.xyz

03Acceptable use

Zones are for content the customer owns or is licensed to distribute. We do not accept open redirectors, credential-harvesting content, malware distribution, or zones whose only purpose is to obscure the origin of unlawful material. Reports of abuse should go to the address above with the full URL and the time of observation; we act on complete reports quickly and ignore incomplete ones.

04Status and changes

Breaking changes to the purge API are announced at least 90 days in advance and old versions stay available for a further year. Non-breaking additions — new response headers, new optional rule fields — ship continuously and are noted in the release summary attached to each edge software version.

Documentation lives here and is versioned alongside the edge software. Start with the overview or go straight to the configuration examples.

Last reviewed: 12 August 2026