Kubernetes is the sometimes-controversial darling of the tech industry. It’s indispensable in situations where it’s useful, but undeniably complex and can be difficult to use. We tend to chalk that up to its relative newness and the complexity of the problems it's solving, but exactly how new are these problems? CI/CD feels new, but its origins predate DevOps. Infrastructure as Code feels new, but we’ve been programmatically defining machines for decades.

Kubernetes isn’t any different; since the dawn of computing, we’ve had issues with workloads and scale. What’s actually going on is that we’re solving the same technical problems over and over again, flavored by the circumstances of the time in which a pain point becomes impossible to ignore. To better understand how Kubernetes came to be, we need to look backwards to its predecessors: the struggle to achieve virtualization, a supercomputer, and a trip to the moon.