Data-Engineering

15 min read

One Namespace, One Workgroup: Datashare Is Redshift Serverless's Back Door

Want a second Redshift Serverless workgroup for isolated compute? We can't attach it to the same namespace—the relationship is 1:1. But there's a back door: create a new namespace/workgroup and use Datashare to access the data from the original namespace. Our new workgroup gets its own compute capacity while still being able to query and JOIN the shared tables.

20 min read

Unlocking the Power of Windowing in Kafka Streams

Windowing refers to the process of dividing a continuous stream of data into discrete segments, or windows, based on time. These windows then serve as the basis for applying computational operations, such as aggregations or transformations, to the data contained within them.