What is the maximum number of consumer groups that can read the same Kafka topic simultaneously?

The Apache Kafka documentation sets no maximum on the number of consumer groups that read the same topic. Topics are multi-subscriber and events can be read as often as needed, because records are not deleted after consumption. Each group tracks its own committed offsets in the __consumer_offsets topic (50 partitions by default), so each group independently reads all the records and has its own lag. The limits you meet in practice are resources: 1) broker CPU and network for the extra fetch requests (caught-up consumers are served from the page cache, lagging ones need disk reads). 2) Offset commit traffic and storage on __consumer_offsets and the group coordinators. 3) Client-side resources. The documented caps are per group, not per topic: group.max.size limits the members of one classic group (default 2147483647). Apache gives no figure for how many groups a cluster can serve, so load-test your own workload.

Expected key concepts: no maximum, multi-subscriber, independently, __consumer_offsets, each group, lag, broker CPU, network, fetch requests, group.max.size, load-test