Kafka Queues promise consumer scaling beyond partitions, but benchmarks show share consumers still lag far behind standard Kafka consumers.
The use case that comes to mind is when the processing time per message is large. Then you'd want to scale-up consumers as the length of the queue grows. Run the benchmark again with a long constant delay in each consumer.
The use case that comes to mind is when the processing time per message is large. Then you'd want to scale-up consumers as the length of the queue grows. Run the benchmark again with a long constant delay in each consumer.