For KStream-to-GlobalKTable joins, what are the output type, windowing, co-partitioning, and join type?

For KStream-to-GlobalKTable joins, what are the output type, windowing, co-partitioning, and join type?

• A. Output: KStream; Windowed; Yes; Key Based
• B. Output: KStream; Non-windowed; No; Key Or Foreign-Key Based
• C. Output: KStream; Non-windowed; Yes; Key Or Foreign-Key Based
• D. Output: GlobalKTable; Non-windowed; No; Key Based

Answer: (B)

Joining a KStream with a GlobalKTable yields a KStream of joined records. This type of join does not use a time window, so it’s non-windowed. The GlobalKTable is replicated across all nodes, so there’s no need for co-partitioning between the stream and the table—the lookup can happen locally on each partition. For how the join key is chosen, you can either use the stream’s key directly or use a foreign key extracted from the value to perform the join.

Joining a KStream with a GlobalKTable yields a KStream of joined records. This type of join does not use a time window, so it’s non-windowed. The GlobalKTable is replicated across all nodes, so there’s no need for co-partitioning between the stream and the table—the lookup can happen locally on each partition. For how the join key is chosen, you can either use the stream’s key directly or use a foreign key extracted from the value to perform the join.