Apples and oranges. DynamoDB is wicked fast for what it does. If you know your use cases ahead of time, you can figure out a data storage strategy where queries

Apples and oranges. DynamoDB is wicked fast for what it does. If you know your use cases ahead of time, you can figure out a data storage strategy where queries and secondary indexes can work wonderfully. $deity help you when the use cases change though. Rearchitect->dump->transform->reload. I really like DynamoDB. Its functional overlap with a fully-featured relational database is very narrow though. No graph queries or joins or advanced data validation or...basically anything on this list: https://www.sql-workbench.eu/dbms_comparison.html https://www.sql-workbench.eu/dbms_comparison.html Postgres can be made to work like DynamoDB: simple queries with no joins, table partitioning and tablespaces for I/O parallelization, etc. DynamoDB barely scratches the surface the other way around. PartiQL is the limit and it's a very basic limit.