Question 1

Which direction does this pairing run?

Accepted Answer

One way only, DynamoDB into MotherDuck. DynamoDB is a source in Polytomic and cannot be written to, so this pairing is about getting key value data somewhere it can be queried in SQL rather than keeping the two in step. If you need results written back to an operational store, that is a different destination rather than something this pairing does.

Question 2

Do I need DynamoDB Streams enabled?

Accepted Answer

For incremental syncing, yes. With Streams on, Polytomic reads changes including deletes, so each run moves only what changed. Without them, every run is a full table scan, which costs read capacity and gets slower as the table grows. There is no tracking field or watermark alternative here the way there is for a relational database, so Streams are the mechanism. Targeting a secondary index can also reduce how much of the table is read.

Question 3

How do maps and lists land in MotherDuck?

Accepted Answer

As JSON. They are decoded recursively so the structure and contents survive, but DynamoDB items carry no detailed schema definition, which means MotherDuck receives them as JSON rather than typed STRUCT and LIST columns. Two things deserve checking after your first load: attributes whose type varies between items, and how number and set representations came through, since those are the ones most likely to need casting in your models.

Question 4

What about a single table design?

Accepted Answer

One physical DynamoDB table arrives as one wide table in MotherDuck. Entity types are not split apart automatically by key prefix or discriminator, so a table holding users, orders, and events together lands as a single table. Splitting them in MotherDuck is usually the easiest path, since you have full SQL there and DuckDB handles the filtering cheaply. Targeting a secondary index that isolates one entity is the alternative if you would rather separate on the way in.

Sync DynamoDB to MotherDuck

Sync DynamoDB to MotherDuck with Polytomic

Streams instead of full scans

Maps and lists decoded

Analysis off the table

Trusted by businesses worldwide

Step-by-step integration guide

Connect DynamoDB in Polytomic

Connect MotherDuck in Polytomic

Create a sync between DynamoDB and MotherDuck

Map fields and enable the sync

Frequently asked questions

The sync and ETL platform that does it all

Powerful data sync

Advanced analytics

Enterprise security

Easy setup

Ready to connect DynamoDB with MotherDuck?