- Market
- Saudi Arabia
- Written for
- CIO, CISO
- Decision
- Architect a cloud estate that keeps regulated data in-Kingdom
- Languages
- EN · AR
The conditions
What a residency claim has to survive across backups, DR and analytics
A residency posture is judged by its weakest copy. These are the conditions an estate has to satisfy before the claim is honest, and each is an architecture decision rather than a setting.
Primary workloads pinned to in-Kingdom regions
Regulated workloads are placed on the regional cloud regions the major providers now operate, and every service they call is pinned to those regions rather than left on a global default.
What it gives you: A home region that is a decision, not an accident of a default.
Backups and replicas enumerated and placed
Snapshot schedules, cross-region replicas and DR copies are listed per workload and pointed at approved regions, because backup services often default to another geography for safety.
What it gives you: No copy created for resilience quietly leaves the boundary.
Telemetry and analytics traced to their storage
Log pipelines, monitoring services and analytics exports are traced to where they actually store data, since observability tooling is the copy teams forget they are making.
What it gives you: The operational exhaust obeys the same boundary as the data.
Evidence maintained before anyone asks for it
The copy trace is kept as a living register per workload, tied to the configuration that enforces it, so the answer to a reviewer is a query rather than an investigation.
What it gives you: A residency answer that takes an afternoon, not a quarter.
Who decides what
The regulator sets the obligation, your compliance team reads it, we build to the reading
It is worth being plain about the chain of authority, because vendors blur it. The obligation comes from the regulator — the NCA’s controls, SAMA’s rules for the institutions it supervises, the PDPL for personal data. What that obligation means for your specific datasets — which classes must remain in-Kingdom, which may cross a border and under what conditions — is an interpretation, and it belongs to your compliance function, not to us and not to a cloud provider’s sales material. Our work starts where that interpretation ends: we take the classification your compliance team signs and architect an estate that demonstrably honours it.
Where the reading is still forming — and for some data classes it is — we architect conservatively, keeping the copy inside the Kingdom when the cost of doing so is small, because a stricter interpretation arriving later should cost you a configuration change, not a migration. The one thing we will not do is settle the regulatory question for you; a supplier who claims to is selling you their opinion dressed as an authority they do not hold.
The copy trace
Every place a cloud workload copies its data without being asked
The primary database is the copy everyone watches. These are the ones that escape a boundary quietly, and the register is where each is caught, pinned and evidenced.
Four kinds of unnoticed copies — automated backups, cross-region replicas, logs and telemetry, analytics exports and caches — converging into a residency register that produces a bounded perimeter, in-scope disaster recovery and evidence on demand.
Automated backups
Snapshot schedules whose target region is a service default nobody chose.
Cross-region replicas
DR copies and read replicas placed for latency or safety, sometimes outside the boundary.
Logs and telemetry
Monitoring and tracing services that store payload fragments wherever the tooling lives.
Analytics exports and caches
Warehouse loads, BI extracts and CDN caches that hold data long after the query.
What we do
The residency register
Every copy enumerated per workload, its region pinned, the enforcing control named.
A bounded data perimeter
The set of places your data can be is finite and known.
DR that stays in scope
Recovery copies protect the business without breaching the boundary.
Evidence on demand
A reviewer’s question is answered from the register, not from memory.
Our role
The residency architecture we carry and the ruling your compliance function keeps
- 01
Classification received, not made
We work from the data classification your compliance team signs off, and when it is incomplete we surface the gap as a question for them rather than filling it with our own judgement.
- 02
The full copy trace
We enumerate every copy each workload makes — backup, replica, log, export — and pin each to an approved region, because the untraced copy is where a residency posture fails.
- 03
Keys and controls where the ruling requires
Encryption, key location and access controls are designed to match the classification, so the technical posture and the compliance position are the same document rather than two.
What the obligation means for your data is your compliance function’s ruling to make; we architect to that ruling, keep the register current as the estate changes, and make the evidence a query rather than a project.
Common questions
What buyers ask
Does hosting in a cloud region inside Saudi Arabia make us compliant by itself?+
No — it settles where the primary copy runs, which is one copy among many. Backups can target another geography by default, monitoring tools store fragments wherever they live, and an analytics export can move a dataset abroad in one job. Compliance is also more than location: access, encryption and the ability to evidence all of it matter to a reviewer. A regional region is the right foundation; the work is tracing everything built on top of it.
Which copies of our data usually leave the Kingdom without anyone deciding it?+
Three recur. Backup and snapshot services whose default target region predates the regional regions, so long-standing schedules still point abroad. Observability tooling — logging, tracing, error reporting — that stores payload fragments in whatever geography the vendor runs. And analytics: a warehouse load, a BI extract or a spreadsheet export made for one meeting that becomes a permanent copy nobody registers. None is malicious; all are defaults. The copy trace exists precisely to catch decisions that were never made.
Can our disaster recovery site sit outside Saudi Arabia if production stays in-Kingdom?+
That depends on the classification of the data in the copy, and it is exactly the kind of question your compliance function rules on — a DR copy is a full copy, so for data that must remain in-Kingdom, the recovery site inherits the same boundary. Architecturally both shapes are workable: the major providers now operate more than one in-Kingdom option, and where a class of data is cleared to leave under conditions, a cross-border DR design can meet them explicitly. What we prevent is the middle case — a DR region chosen for convenience that quietly contradicts the posture the rest of the estate maintains.
In this cluster
Next step
Scope it with us
A residency mapping session that ends with the copies your estate makes today, which of them leave the Kingdom, and the order in which to bring them back.

