A company has a 24 TB MySQL database in its on-premises data center that grows at the rate of 10 GB per day. The data center is connected to the company’s AWS infrastructure with a 50 Mbps VPN connection.
The company is migrating the application and workload to AWS. The application code is already installed and tested on Amazon EC2. The company now needs to migrate the database and wants to go live on AWS within 3 weeks.
Which of the following approaches meets the schedule with LEAST downtime?

A.
1. Use the VM Import/Export service to import a snapshot of the on-premises database into AWS.
2. Launch a new EC2 instance from the snapshot.
3. Set up ongoing database replication from on premises to the EC2 database over the VPN.
4. Change the DNS entry to point to the EC2 database.
5. Stop the replication.

B.
1. Launch an AWS DMS instance.
2. Launch an Amazon RDS Aurora MySQL DB instance.
3. Configure the AWS DMS instance with on-premises and Amazon RDS database information.
4. Start the replication task within AWS DMS over the VPN.
5. Change the DNS entry to point to the Amazon RDS MySQL database.
6. Stop the replication.

C.
1. Create a database export locally using database-native tools.
2. Import that into AWS using AWS Snowball.
3. Launch an Amazon RDS Aurora DB instance.
4. Load the data in the RDS Aurora DB instance from the export.
5. Set up database replication from the on-premises database to the RDS Aurora DB instance over the VPN.
6. Change the DNS entry to point to the RDS Aurora DB instance.
7. Stop the replication.

D.
1. Take the on-premises application offline.
2. Create a database export locally using database-native tools.
3. Import that into AWS using AWS Snowball.
4. Launch an Amazon RDS Aurora DB instance.
5. Load the data in the RDS Aurora DB instance from the export.
6. Change the DNS entry to point to the Amazon RDS Aurora DB instance.
7. Put the Amazon EC2 hosted application online.