Sumeta — the company behind Bridge
Question · residency

How do you meet data residency requirements without a region lock-in?

Residency is usually sold as a region selector — a dropdown that moves the vendor's copy of your data to a different country. Structural residency is different: the data and the metadata never leave infrastructure you control, so there is no vendor-side region to choose.

When both the data plane and the control plane run in your environment, residency is a property of your architecture rather than a vendor setting. There is no cross-border copy to audit because no vendor-side copy exists.

0
Vendor-side copies of rows or schema metadata
0
Inbound connections required
4/4
Data flows kept inside your perimeter
1
Deployment boundary you can show an auditor
The structural argument

Region selection vs structural residency

These are not two flavours of the same thing. One is a contract about where a vendor keeps data; the other is an architecture in which the question does not arise.

Two models, one question: where does the data actually go?
PropertyRegion-selectable SaaSStructural residency (Bridge)
Where rows are processedVendor's cloud, in the chosen regionYour VPC, on-prem or isolated network
Where metadata livesVendor's control planeYour deployment
What crosses a borderRows, metadata and logs, by designNothing, unless you target an external system
How you prove itVendor attestation and DPAYour own network and packet evidence
Effect of a new regulatorRe-negotiate region availabilityBoundary is already inside your control

This matters commercially as well as legally. A region dropdown ties you to whichever regions the vendor has launched, and to their launch schedule. A structural boundary ties you to infrastructure you already operate. If your requirement is "data must not leave our environment", a region selector is answering a different question.

For teams that do need to sync across a boundary, Bridge supports file exchange across a security gateway: the change set is carried across deliberately, by your rules, with no inbound connection opened.

You need structural residency when…

  • ✓The requirement is "data must not leave our environment", not "must stay in country X"
  • ✓Your auditor asks for network evidence rather than a vendor letter
  • ✓You operate multiple jurisdictions and want one consistent control
  • ✓You cannot wait for a vendor to launch a region in your market

Structural residency is overkill when…

  • ×You are comfortable with a vendor-hosted control plane
  • ×Your only requirement is a contractual region commitment
  • ×You have no infrastructure to host the platform on
  • ×You need a fully managed service with no operational burden

Residency, answered

Do you make specific country or region commitments?

We deliberately do not sell residency as a list of regions. Bridge runs in your environment, so the boundary is your infrastructure and your jurisdiction — there is no vendor-side region holding a copy of your data to make a promise about.

What crosses the network boundary?

Only what you configure. Source reads and target writes happen inside your perimeter; for isolated segments a change set crosses a security gateway via file exchange, under your rules. No inbound connection is required.

How do I verify it during a POC?

Instrument the egress interface, run a pipeline, and confirm that no flow carries rows, schema metadata or logs to a Sumeta endpoint. The control plane runs locally, so there is no vendor console to fall back on.

Does this replace a DPA?

No. Enterprise includes DPA and audit support, and the DPA still governs the commercial relationship. Structural residency simply means the data-handling surface is much smaller.

Prove residency in your own environment

Run a pipeline with egress instrumentation and confirm the boundary yourself before you sign.