Custom Solution or ETL?

Hi, there are multiple option available for your use cases. AWS Glue: it could do the sync job but, as you mention, given the size of the data that need to be handled I would recommend the following options:

• An AWS Lambda Function with an Amazon CloudWatch Trigger (https://docs.aws.amazon.com/AmazonCloudWatch/latest/events/RunLambdaSchedule.html) that will run periodically and it will copy the data from database A to database B. Lambda Function also support Java and with the new feature Lambda SnapStart (https://docs.aws.amazon.com/lambda/latest/dg/snapstart.html) it can easily support SpringBoot with good performance. Important notice, you should be able to complete your sync task within 15 minutes.
• Using Amazon DMS (https://aws.amazon.com/dms/). If the databases that you are talking about is from the supported DMS list (https://docs.aws.amazon.com/dms/latest/userguide/CHAP_Source.html) Amazon DMS will completely handle the sync for you without need to write any code. Amazon DMS will perform the sync using CDC task (https://docs.aws.amazon.com/dms/latest/userguide/CHAP_Task.CDC.html) that is designed for continuous replication of the data (depending your requirement a continuous replication mechanism could not be needed).

To conclude, given that Amazon DMS with a CDC task is meant to be used for continuous replication, it is likely to be more expensive of a solution running on Lambda Function triggered with low frequency; for this reason I would recommend to use the AWS Lambda approach. https://aws.amazon.com/lambda/pricing/ https://aws.amazon.com/dms/pricing/