Regions
Active regions, the hub and compute-edge model, regional URLs, and placement semantics
Active regions
Scalix Cloud runs in two regions, both in the EEA. The live catalog is served by GET /v1/regions; the table below matches it as of November 2026.
| Region | Location | Role | Services |
|---|---|---|---|
eu-central-1 | Falkenstein, Germany (hetzner-fsn1) | Full hub | Run, Functions, Computers, Sandboxes, ScalixNova databases, object storage |
eu-north-1 | Helsinki, Finland (hetzner-hel1) | Compute edge | Run, Functions, Computers, Sandboxes |
Hub and compute edge
eu-central-1 is the full hub: every service runs there. eu-north-1 is a compute edge: workloads boot and serve locally, while databases and object storage stay in eu-central-1. Both regions are in the EEA, so data remains within the EU.
Control-plane services such as the AI router run at hub points of presence only, never per region. Edge nodes forward to the hub over a private WireGuard mesh, so guests on an edge reach hub services without traversing the public internet.
Regional URLs
Run and Functions expose regional hostnames:
https://{name}.{region}.run.scalix.world
https://{name}.{region}.fn.scalix.worldFor example, myapp.eu-north-1.run.scalix.world terminates on the Helsinki edge. The apex forms *.run.scalix.world and *.fn.scalix.world (no region label) remain served by eu-central-1.
Choosing a region
Placement is controlled at two levels:
data_region(organization): the default region for new resources in that org.region(per resource): a soft preference. The scheduler tries to honor it but may place elsewhere when capacity requires.region_pinned: true(per resource): a hard pin. Placement fails closed when the pinned region has no capacity; the resource is never silently placed in another region.
The list of selectable regions in the console and API is driven by GET /v1/regions. New regions are added to that catalog without client changes.