Roger Hoover, one of the first engineers to work on Confluent Cloud, joins Tim Berglund (Senior Director of Developer Experience, Confluent) to chat about the evolution of Confluent Cloud, all the stages that it’s been through, and the lessons he’s learned on the way. He talks through the days before Confluent Platform was created, and how he contributed to Apache Kafka® to run it on OpenStack (the feature used to separate advertised hostnames from the internal hostnames).
The Confluent Cloud control plane is now run in over 40 regions. Under the covers, Roger and his team are managing tens of thousands of resources at the cloud provider layer. This means creating VPCs, VMs, volumes, and DNS records, to manage software artifacts, like what version of Kafka is running and user management. Confluent Cloud is a complex application and distributed system spread across the entire world, but Roger reveals how it's done.