ARA-C01 Question 116
Select 2Domain 2.0: Snowflake Architecture (30%)A global software company runs a single Snowflake account in AWS us-east-1 for all analytics workloads. The company is acquiring a business unit in Europe that must keep its customer data physically in the EU for regulatory reasons. The architecture team also wants to share a curated product usage dataset between the US and EU environments with minimal data movement and low operational overhead. Which TWO actions should the architect recommend to meet these requirements?
- A
Create a separate Snowflake account in an EU region and use replication/failover features to keep all tables synchronized between the US and EU for shared analytics
- B
Create a separate Snowflake account in an EU region for regulated workloads, and use Secure Data Sharing or listings where supported across regions/clouds to provide the curated dataset without copying data into files
- C
Keep a single Snowflake account in AWS us-east-1 and enforce EU residency by storing EU customer rows in separate schemas with masking policies
- D
Use database replication to migrate the regulated EU customer data into the US account for centralized querying, because replication preserves the original data residency classification
- E
Deploy the EU business unit in a Snowflake account located in an EU region, and design data products so only the specific curated datasets that are allowed for cross-region consumption are shared to the US environment
Show answer and explanation
Correct answers: B, E
Explanation
Snowflake data residency is determined by the cloud platform and region where an account is deployed. If data must remain physically in the EU, the architect should provision a Snowflake account in an EU region for those workloads rather than rely on logical controls inside a US-based account. For collaboration across regions, Snowflake-native sharing capabilities are generally preferred over ETL/export pipelines because they reduce operational overhead and avoid file-based duplication. However, architects must still ensure that only approved datasets are shared and that the chosen sharing pattern is supported for the involved regions/clouds and compliant with regulatory requirements. Replication and failover are valuable for disaster recovery and business continuity, but they create additional physical copies and are not the right default approach when the goal is to minimize data movement and maintain strict regional boundaries. These recommendations align with Snowflake best practices for account strategy, region selection, data sharing, and cross-region architecture.
- A. Incorrect.
Incorrect. Creating an EU account is necessary for EU data residency, but using replication/failover to keep all tables synchronized between the US and EU does not align with the requirement to minimize data movement and preserve regulatory boundaries. Replication creates physical copies of database objects in another region/account. Replicating all tables would likely move regulated data across regions, which is exactly what the company is trying to avoid except where explicitly permitted.
- B. Correct.
Correct. A separate Snowflake account in an EU region is the right architectural choice for data that must remain physically in the EU. For sharing a curated dataset between environments, Secure Data Sharing is the preferred Snowflake-native mechanism because it provides live access to shared data without unloading to files or building custom pipelines. In cases where sharing spans regions or clouds, listings/cross-region sharing patterns are the lower-overhead design compared with ETL-based duplication, assuming the specific sharing configuration is supported and compliant.
- C. Incorrect.
Incorrect. Data residency is enforced by where the Snowflake account is deployed, not by logical separation such as schemas, row filters, or masking policies inside an account located in a non-EU region. Masking policies protect sensitive values from unauthorized viewers, but they do not change the physical location where the data is stored. Keeping EU-regulated data in a US-hosted account would violate the residency requirement.
- D. Incorrect.
Incorrect. Database replication creates copies in target accounts/regions for business continuity and read-only access scenarios, but it does not preserve EU residency by some metadata label while physically moving data to the US. If regulated data must stay in the EU, replicating it into a US account is not an appropriate control. This option reflects a common misconception that governance tags or classifications can substitute for physical residency controls.
- E. Correct.
Correct. The core architectural requirement is to place regulated workloads in a Snowflake account deployed in an EU region. Then, data sharing should be deliberate and limited to approved curated datasets rather than broad replication of source systems. This reduces operational overhead, limits unnecessary data movement, and follows least-privilege/data-minimization principles. It is the most practical design for satisfying both residency and cross-region collaboration requirements.