Apache Kafka Stretch Cluster Deployment Guide

This guide details the planning and configuration steps for deploying an Apache Kafka stretch cluster, particularly in a three Data Center (DC) setup. It begins by outlining the reference architecture required for such a deployment, emphasizing that the architecture is primarily intended for on-premises installations, while also noting considerations for cloud deployments. The guide explains the importance of naming the Data Centers and how this affects the configuration of Kafka brokers. It discusses the implications of choosing the number of replicas within each DC, including how this choice impacts availability and data durability. The document also covers the necessary calculations for replication factors and minimum in-sync replicas, ensuring that the cluster can withstand DC failures. Furthermore, it addresses the need to collect network information to calculate buffer sizes, highlighting the importance of managing latency in a stretch cluster environment. Finally, the guide provides instructions for configuring Kafka brokers to ensure effective deployment.