Migrate from Amazon RDS for Oracle to Amazon RDS for MySQL

Validate the source and target database versions and engines.

Identify hardware requirements for the target server instance.

Identify storage requirements (storage type and capacity).

Choose the proper instance type (capacity, storage features, network features).

Identify network-access security requirements for the source and target databases.

Choose an application migration strategy.

Consider whether you want full downtime or partial downtime for cutover activities.

Create a VPC and subnets.

Create security groups and network access control lists (ACLs).

Configure and start the Amazon RDS for Oracle instance.

Configure and start the Amazon RDS for MySQL instance.

Prepare a test case for validation of code conversion.

This will help in unit-testing for the converted code.

Configure the AWS DMS instance.

Configure source and target endpoints in AWS DMS.

Generate the target database script using AWS SCT.

Check the accuracy of the code that was converted by AWS SCT. Some manual work will be required.

In AWS SCT, choose the "Case Insensitive" setting.

In AWS SCT, choose Project Settings, Target Case Sensitivity, Case Insensitive.

In AWS SCT, choose not to use the Oracle native function.

In Project Settings, check the functions TO_CHAR/TO_NUMBER/TO_DATE.

Make changes for "sql%notfound" code.

You might have to convert the code manually.

Query on tables and objects in stored procedures (use lowercase queries).

Create the primary script after all changes are made, and then deploy the primary script on the target database.

Unit-test stored procedures and application calls using sample data.

Clean up data that was created during unit testing.

Drop foreign key constraints on the target database.

This step is required to load initial data. If you don't want to drop the foreign key constraints, you must create a migration task for data specific to the primary and secondary tables.

Drop primary keys and unique keys on the target database.

This step results in better performance for the initial load.

Enable supplemental logging on the source database.

Create a migration task for the initial load in AWS DMS, and then run it.

Choose the option to migrate existing data.

Add the primary keys and foreign keys to the target database.

Constraints need to be added after the initial load.

Create a migration task for ongoing replication.

Ongoing replication keeps the target database synchronized with the source database.

Replace Oracle native functions with MySQL native functions.

Make sure that only lowercase names are used for database objects in SQL queries.

Shut down the application server.

Validate that the source and target databases are in sync.

Stop the Amazon RDS for Oracle DB instance.

Stop the migration task.

This will stop automatically after you complete the previous step.

Change the JDBC connection from Oracle to MySQL.

Start the application.

Review and validate the project documents.

Gather metrics about time to migrate, percentage of manual versus tool tasks, cost savings, etc.

Stop and delete AWS DMS instances.

Remove the source and target endpoints.

Remove migration tasks.

Take a snapshot of the Amazon RDS for Oracle DB instance.

Delete the Amazon RDS for Oracle DB instance.

Shut down and delete any other temporary AWS resources you used.

Close the project and provide any feedback.