The requirements emphasize minimum downtime, rapid migration, managed database services, high availability, and support for stored procedures. For Oracle-to-AWS migrations with minimal downtime, AWS DMS can replicate ongoing changes from the source database to the target, which reduces cutover downtime. However, a critical constraint is that the database contains stored procedures. If the company moves to a different engine, it must convert schema objects and procedural code.
Option D is the only option that combines an online replication approach (AWS DMS) with a managed, highly available target database (Amazon Aurora PostgreSQL) and explicitly includes schema and stored procedure conversion with AWS SCT. This aligns with a fast migration approach: SCT prepares the converted schema and stored procedures, while DMS performs ongoing replication of data changes until cutover, minimizing downtime. Aurora PostgreSQL is a managed service and supports high availability through multi-AZ architecture and managed failover capabilities within a Region.
Option A is incorrect because Amazon RDS does not use an S3 bucket “as database storage” in the way described. RDS storage is managed by the RDS service, and S3 replication is not the mechanism used to provide database high availability. High availability for RDS is typically achieved through Multi-AZ configurations and/or cluster-based architectures depending on engine. The described storage approach is not a valid managed RDS architecture.
Option B does not meet the managed database requirement because it restores Oracle onto EC2 instances, which is self-managed. It also does not meet minimum downtime because it requires shutting down the on-premises database to avoid new transactions before restoring the backup. This is a longer outage model than an online migration with continuous replication.
Option C is not feasible and would require major redesign. DynamoDB is not a relational database replacement for Oracle stored procedures. DAX is a DynamoDB cache and cannot run stored procedures. The option also incorrectly suggests running stored procedures “in DAX.” This does not meet the “migrate as quickly as possible with minimum downtime” requirement due to extensive refactoring and redesign.
Therefore, migrating to Aurora PostgreSQL using AWS SCT for schema and stored procedure conversion, and AWS DMS for ongoing replication to minimize downtime, is the best fit.
[References:AWS documentation and best practices describing AWS DMS as a tool for minimal-downtime database migration through continuous replication.AWS documentation describing AWS SCT for converting schema objects and procedural code when migrating between different database engines.AWS documentation describing Amazon Aurora as a managed, highly available database service with built-in multi-AZ resilience.]
