Certified Kubernetes Administrator (CKA) Study Guide

Chapter 6. Storage

When container images are instantiated as containers, the container needs context—context to CPU, memory, and I/O resources. Pods provide the network and the filesystem context for the containers within. The network is provided as the Pod’s virtual IP address, and the filesystem is mounted to the hosting node’s filesystem. Applications running in the container can interact with the filesystem as part of the Pod context. A container’s temporary filesystem is isolated from any other container or Pod and is not persisted beyond a Pod restart. The “Storage” section of the CKA curriculum addresses the technical abstraction in Kubernetes responsible for persisting data beyond a container or Pod restart.

A volume is a Kubernetes capability that persists data beyond a Pod restart. Essentially, a volume is a directory that’s shareable between multiple containers of a Pod. You will learn about the different volume types and the process for defining and mounting a volume in a container.

Persistent volumes are a specific category of the wider concept of volumes. The mechanics for persistent volumes are slightly more complex. The persistent volume is the resource that actually persists the data to an underlying physical storage. The persistent volume claim represents the connecting resource between a Pod and a persistent volume responsible for requesting the storage. Finally, the Pod needs to claim the persistent volume and mount it to a directory path available to the containers ...