We have a couple of mySql tables in RDS that are huge (over 700 GB), that we'd like to migrate to a DynamoDB table. Can you suggest a strategy, or a direction to do this in a clean, parallelized way? Perhaps using EMR or the AWS Data Pipeline.
You can use AWS Pipeline. There are two basic templates, one for moving RDS tables to S3 and the second for importing data from S3 to DynamoDB. You can create your own pipeline using both templates.
Regards