Is it possible to overlap duplicates when engaging in DynamoDB cross region table copying?

Question

I'm trying to set up a pipeline to copy a table between two Dynamo DB geographic servers and I can't find a way to explicitly make incremental copies every two hours between a single table that exists on Amazon's East coast server to the West coast server overlapping duplicates when copying. The job runs every two hours however looking through the documentation for Amazon Data Pipeline I can't find anything to allow overwrites if the record already exists on the West coast table.

Is there any way to do it through Amazon's interface or is this the default behavior? I'm looking for a way to be sure that inserting the same hash and range key will overwrite the existing record during an incremental copy.

Answer 1

The Amazon Data Pipeline uses the Hive Connector. You can see this by looking at the details of your pipeline.

I believe the Hive Connector uses PutItem. You should be able to confirm this by looking at your DynamoDB cloudwatch graphs and look at the N stat of the PutItem and UpdateItem item latency metrics. In this case, it is just overwriting what is already there. The hive connector does not allow you to do conditional puts, as far as I know, which is how you would check that the item already exists before putting it.