Ever wondered how far Kafka can really go before it breaks?

In this hands-on session, we’ll explore Kafka’s performance under massive load — millions of messages per second — and learn how to monitor, tune, and stabilize it in real time.

You’ll learn how to:

- Design high-throughput topics and partitions for parallel load

- Use Kafka’s native performance tools and k6 to generate realistic producer/consumer traffic

- Measure and visualize throughput, latency, and consumer lag with Prometheus + Grafana

- Detect bottlenecks across network, disk, and broker configurations

- Apply tuning strategies for batching, compression, and replication to reach optimal throughput

We’ll build the setup live, crank up the load, and watch Kafka’s metrics under stress — learning exactly what happens when a distributed system meets its limits.

Perfect for developers, DevOps engineers, and architects who want to understand Kafka’s real-world scalability and performance tuning.