ADA-C01 Question 509
Select 3Replicate database objectsA global company uses Snowflake Business Critical edition in AWS us-east-1 for production. The administration team must maintain a read-only disaster recovery copy of a critical database in AWS us-west-2. The source database contains permanent and transient tables, secure views, streams, tasks, and an internal stage used by ETL jobs. The team wants a solution that minimizes manual rebuild work after failover and keeps the secondary database synchronized regularly. Which actions should the Snowflake administrator take to meet these requirements?
- A
Enable replication for the account, create the target account in the same organization, and use database replication with scheduled refreshes to keep a secondary database in sync.
- B
Use database cloning from the primary account into the secondary account on a schedule, because clones automatically stay synchronized after the initial copy.
- C
Plan to recreate streams and tasks after failover, because these object types are not replicated as usable objects in the secondary database.
- D
Rely on replication of the internal stage contents, because all data files stored in internal stages are included automatically with database replication.
- E
Configure the replicated database in the target region as a secondary database and promote it during a disaster event when write access is required.
Show answer and explanation
Correct answers: A, C, E
Explanation
The best answer is to use Snowflake database replication and maintain the target as a secondary database that is refreshed regularly, then promoted during a disaster. This is the standard Snowflake approach for cross-region disaster recovery of database objects. Candidates should know that replication is not the same as cloning: clones are point-in-time copies and do not stay synchronized. They should also understand object behavior nuances during replication. In particular, administrators should not assume every object behaves identically after failover; operational objects such as streams and tasks require recovery planning. Likewise, stage file handling is a common source of confusion and should be validated explicitly in a DR design rather than assumed. Snowflake documentation on replication and failover groups, database replication, primary/secondary databases, and replicated object considerations provides the relevant guidance and best practices.
- A. Correct.
Correct. Cross-region database replication is the appropriate Snowflake capability for maintaining a disaster recovery copy in another region. Replication requires supported editions/features and accounts that are part of the same organization. A secondary database can be refreshed on a schedule or manually, which satisfies the requirement to keep the copy synchronized with minimal manual effort.
- B. Incorrect.
Incorrect. Database cloning does not create a continuously synchronized copy across accounts/regions for disaster recovery. Cloning is a point-in-time metadata operation and does not provide ongoing replication. This option reflects the common misconception that clones behave like replicated mirrors.
- C. Correct.
Correct. Streams and tasks have special behavior with replication. While their definitions can be included in replication, they are not replicated as actively usable runtime objects in the secondary for seamless continued processing. In practice, administrators should plan operational recovery steps for these objects after failover, such as recreating or re-enabling them as appropriate.
- D. Incorrect.
Incorrect. Internal stage contents are not fully protected by assuming database replication will include all staged files automatically. A common mistake is to treat stage data like table data. Administrators should not rely on database replication alone to preserve all internal stage file contents for DR planning.
- E. Correct.
Correct. In Snowflake disaster recovery design, the target copy is maintained as a secondary database. If the primary region becomes unavailable and the organization needs to resume write operations, the administrator promotes the secondary database. This aligns with Snowflake's primary/secondary database replication model.