Some of us developers still care about low-latency workloads. And for those, traditional Apache Kafka is still king (I know—_"but RedPanda"—_depending on the workload, Kafka can do better). For those of us who do need latency, and care about optimizing systems, this is a fantastic read: https://blog.allegro.tech/2024/03/kafka-performance-analysis.html

For all these distributed data systems, finding the bottlenecks are on a well configured systems is already a pretty tough job. It gets a lot worse on poorly configured systems, of course. It's good to have those snapshots though, but they sure do get out of date quickly too. I'm not following Red Panda development very closely, but I wouldn't be surprised if one or two of their people chimed in on this.