Infrastructure & Isolation

Full Fabric runs every institution on one shared platform: the same application release, the same servers and the same load balancing. Each institution is a separate tenant. Its data lives in a database of its own, and every request, background task and integration call is tied to exactly one institution before it touches any data.

This page explains how that separation works. For the controls around who can see what inside an institution, see Security & Compliance and Identity & Auth.

How a Request Finds Its Institution

Every institution has a handle, a short lowercase name such as ff-business-school. The handle gives the institution its web address:

https://ff-business-school.fullfabric.cloud

Full Fabric reads the address of each incoming request and works out which institution it belongs to:

  • An address that does not match an institution gets a not found response. No data is read.
  • A request is connected to one institution's database only. It cannot read or write another institution's data on the way.
  • Sign-in and permission checks happen after the institution is known, so a user account only exists, and only works, inside its own institution.

Institutions can also serve the platform on their own domain, for example apply.example.com. See Custom Domains.

What Each Institution Has of Its Own

Area How it is kept apart
Data Profiles, applications, offers, payments, forms, emails and every other record live in the institution's own database
Configuration Settings, modules and features, roles and teams, schemas and custom fields, lifecycle states, email templates and branding are stored in that same database
User accounts Staff, applicant and student accounts belong to one institution. The same email address at two institutions means two separate accounts
API tokens A token is created in one institution and only works on that institution's addresses
Background work Queued work, such as sending emails, running imports and building exports, is stored in the institution's database and processed while connected to that institution only
Search Global search runs directly against the institution's own database. There is no shared search index
Data warehouse Each institution has its own warehouse and its own read-only credentials. See Data Warehouse
Sandboxes Each sandbox is a further database of its own, separate from production. See Sandboxes

What Is Shared

All institutions run on the same application. Shared parts carry no institution data of their own:

  • Application servers and load balancers handle requests for every institution and connect each request to the right database.
  • The staff and portal frontends are the same code for everyone, served from a content delivery network.

Which areas an institution can use is decided by its enabled modules and features, not by a separate installation. See Features & Permissions.

Note: A shared application does not mean shared data. Each request reads and writes only the database of the institution it belongs to.

Sandboxes and Isolation

A sandbox is a full, separate environment linked to its production institution:

  • It has its own database, created empty from default settings. No production data is copied into it.
  • Work done in a sandbox never changes production, and production changes never reach a sandbox.
  • Deleting a sandbox removes its database and everything in it.

See Sandboxes for how to create, activate and delete them.

Isolation Inside an Institution

Separating institutions is the first layer. Inside an institution, Full Fabric limits what each person can reach:

Control What it limits Read more
Roles and features Which areas and actions a user can use Features & Permissions
Data access scopes Which programmes, classes and courses a staff member can see Data Access Scopes
Schema permissions Which fields a user can view or edit Schema Permissions
Portal access Applicants and students see only their own records in the portal User Types & Roles

Frequently Asked Questions

Can staff at one institution see another institution's data?

No. A staff account exists only inside its own institution, and every request is connected to that institution's database alone.

Do two institutions in the same group share data?

No. Each institution is a separate tenant with its own database, even when several belong to the same university group.

Does a custom domain change where data is stored?

No. A custom domain is another address for the same institution. The data stays in the same database. See Custom Domains.