Author: Arvind Rajagopal

A global investment bank and Confluent used Apache Kafka to deliver sub-5ms p99 end-to-end latency with strict durability. Through disciplined architecture, monitoring, and tuning, they scaled from 100k to 1.6M msgs/s (<5KB), preserving order and transparent tail latency.

Sharing metadata on the data you store in your Confluent cluster is paramount to allow for effective sharing of that data across the enterprise. As the usage of real-time data […]