Kubernetes has many core abstractions, typically known as primitives, that make the expertise of deploying and managing functions so a lot better than what got here earlier than. Understanding these abstractions helps you’re taking full benefit of Kubernetes and likewise keep away from complexity—particularly when working stateful functions like databases, knowledge analytics, huge knowledge functions, streaming engines, machine studying, and AI apps.
In this text, I’ll overview among the elementary abstractions in Kubernetes storage, and stroll by means of how Portworx PX-Enterprise helps resolve vital challenges that come up with the necessity for persistent storage in Kubernetes.
Kubernetes abstractions and Kubernetes storage
The Pod is a superb instance of a core Kubernetes abstraction. It’s really the primary instance—the place to begin.
Back in 2015, different container orchestration methods began with a single container as the elemental abstraction; Kubernetes began with Pods. A Pod is a bunch of a number of containers that must run collectively to be helpful. One easy analogy is {that a} Pod is like an outfit of clothes. It’s nice to have a shirt and socks on, however let’s not stroll out the door with out pants!
Pods are like that—they allow us to deal with what’s wanted to be helpful (a working outfit) and never overload us with bookkeeping trivialities (a shoelace, one sock). Don’t get me unsuitable, the trivialities is being tracked by the scheduler and Kubelet (the Kubernetes agent). But it’s this abstraction that enables for the ecosystem to construct on Kubernetes and for directors to automate their infrastructure. And right now, we see that almost all different schedulers have adopted the Pods idea, a certain signal of its usefulness.
