A region is a hardware set the Stack Harbor PaaS installation exposes for hosting environments. Each region has its own capacity, location, and characteristics, but they are all orchestrated through the same platform — you operate them through one dashboard, one CLI, and one API.
When an installation exposes more than one region, every new environment carries a region choice. This page covers how to make that choice and where to find the details about each region before you commit.
What region choice affects
- Geographical location — pick the region closest to your users for the lowest response time, or geographically separated regions for redundancy.
- Capacity and quality of hardware — different regions can have different hardware tiers. A common pattern is to run development and testing on lower-spec regions and production on higher-spec ones.
- Available features — some features (live migration between data centres, certain database HA topologies) are only available within or between specific regions.
- Environment domain — the auto-assigned domain name may vary per region, depending on installation configuration.
Where region selection appears
Every flow that creates an environment exposes a region drop-down at the top of the dialog:
- The topology wizard.
- The platform marketplace.
- The environment import dialog.
The one flow that does not ask is cloning. A clone is created in the same region as the source environment automatically. Once the clone exists, you can migrate it to a different region using the migration option if needed.
After an environment is created, its region is identified by a small icon next to the environment in the dashboard. Hover the icon to see the region name.
Reading region details
Each region in the selector is annotated with a short description inline. For the full picture, expand the More details entry — it opens the Regions tab of the installation’s Quotas & Pricing window, where you can read:
- General description and the list of supported features for each region.
- Available resource limits, charged resources (RAM/CPU, disk, traffic, options), and licensing breakdowns per region.
- Any region-specific notes the installation operator has published.
If your installation operator has published a separate document or website for region details, the More details link points there instead.
How to choose
A few rules of thumb that work in practice:
- Latency-bound workloads — pick the region geographically closest to where the traffic originates.
- Compliance-bound workloads — pick the region in the legal jurisdiction your data must stay in.
- HA and DR — pick two regions, ideally in separate data centres. Then deploy two environments (one per region) and front them with the appropriate routing layer.
- Dev and prod parity — pick the same region for both if you can, so your latency profile in dev matches production.
- Cost-driven choice — read the region’s resource and licensing breakdown before committing. Per-region pricing can differ.
Changing your mind
You are not locked in. If the region you picked is no longer the right one, you can move the environment without recreating it. See environment region migration — both live (zero-restart) and offline migration are supported, depending on the source and target regions.