KRaft dramatically improves metadata scalability. ZooKeeper-based Kafka could handle ~200k partitions before controller failover became too slow (20-30s). KRaft supports millions of partitions and reduces controller failover to under 1 second. Metadata propagation from controller to brokers is also faster because brokers actively replicate the metadata log rather than receiving individual RPC calls.
Lesson 5 of 5
KRaft vs ZooKeeper: Performance
Try it in the simulator
Guided scenarios for this module — the simulator opens with the exercise already running.