Designing Topic Structures for Data Resiliency and Disaster Recovery

You've analyzed your business continuity and disaster recovery requirements and designed a comprehensive multi-cluster Kafka architecture to meet your overall goals. What's next? In this talk, we'll discuss the actual implementation details for the clients and topics that live in multi-cluster environments, including:

What naming conventions and patterns should be followed for topics in a multi-cluster architecture? How does this differ between application and internal topics? Which topics should be replicated between clusters and which ones should live only in their respective cluster(s)? How should clients of our platform interact with their topics, and how is this different for different types of clients (producers, consumers, Kafka Streams, and Kafka Connect)?

What do failover and failback look like in different scenarios? Our goal is to develop a better understanding of the considerations that go into implementing a successful multi-cluster architecture.