ARA-C01 Question 135
Single answerSharing across cloud regionsA global data provider maintains a Snowflake account in AWS us-east-1 and publishes curated datasets to external customers using secure shares. A new customer requires access from their Snowflake account in Azure West Europe and does not want to build ETL pipelines or manage file exports. The provider wants the customer to query the shared data with minimal operational overhead while keeping the provider's existing AWS-based data publishing process. Which solution should the architect recommend?
- A
Create a database replication group that includes the provider's source database, replicate it to the customer's Azure region, and create a secure share from the replicated database in the target region.
- B
Add the customer's Azure account directly to the existing secure share in AWS us-east-1 because secure sharing supports cross-cloud and cross-region access without replication.
- C
Unload the shared tables to cloud storage in AWS, provide the files to the customer, and let the customer create external tables in Azure over the exported data.
- D
Create a data exchange listing from the existing share; listings automatically provide live cross-cloud access to any consumer account without additional regional setup.
Show answer and explanation
Correct answer: A
Explanation
For Snowflake-native sharing, architects must distinguish between standard secure data sharing and cross-region/cross-cloud sharing requirements. Standard secure shares are straightforward when provider and consumer are in compatible locations, but when a provider in AWS us-east-1 needs to share with a consumer in Azure West Europe, the provider typically needs to replicate the relevant database objects to the target region/cloud and then create the share there. This lets the consumer query shared data natively in Snowflake without ETL pipelines, file exports, or external table complexity. Data Exchange and listings improve discoverability and governance of shared datasets, but they do not remove the need to satisfy the underlying region/cloud data availability requirement. This aligns with Snowflake documentation and best practices for replication and secure data sharing across regions and clouds.
- A. Correct.
Correct. Secure sharing is native within the same region/cloud, but direct sharing does not by itself provide live access across different clouds/regions. For cross-region or cross-cloud scenarios, the provider must use Snowflake replication to make the database available in the target region, then share from that replicated database to the consumer account in that region. This approach preserves the provider's existing Snowflake-based publishing model and avoids ETL/file-based distribution.
- B. Incorrect.
Incorrect. This reflects a common misconception that secure shares are globally accessible from any Snowflake account. In reality, direct secure sharing is constrained by region/cloud locality. Cross-region and cross-cloud data sharing requires replication/failover groups or a supported cross-region sharing architecture, not simply adding a remote account to the existing share.
- C. Incorrect.
Incorrect. Although this could technically move data to the customer, it introduces exactly the kind of operational overhead the scenario wants to avoid: exports, file management, and customer-side table definitions. It is not the Snowflake-native sharing approach best suited for low-maintenance cross-region distribution.
- D. Incorrect.
Incorrect. Listings and exchanges help with discovery and distribution workflows, but they do not eliminate the underlying regional/cloud constraints. A listing does not automatically grant live cross-cloud query access to data that has not been replicated or otherwise made available in the consumer's region.