Here you find a selection of definitions for containerization technology. You should read about Understanding Docker first.
An image can be seen as an operating system snapshot. It has all dependencies, a piece of software needs, preinstalled. It can be easily shared and thus allows reproducible environments for code to run in.
A container is a running an image. A container normally is only running one Application. For example, in a standard web application, you would have one container running your frontend, one running the backend and one for the Database. There can be multiple containers running the same image.
The host system is the system where the container is running in.
A volume can be mounted into a container to get data persistency in an otherwise volatile environment. A volume can be imagined as a USB-stick that you plug into your container, every time you start it. Even if you kill the container and build a new one, the USB-stick can stay the same and your data is persited.
A container runtime is responsible for actually running a container on the host machine. Docker provides a container runtime, but there are also standalone runtimes (e.g. Containerd, which is used by Kubernetes).
A container registry is a repository or database where container images are saved. These images can be public, like the Docker Hub, or private, like the integrated GitLab Registry. When running a container, you can specify where an image should be pulled from. It is a best practice to build an image in a CI pipeline, push it to a registry and while deploying pull the image again.
Read how to set up your Kubernetes Cluster to pull from the gitlab registry.
Docker itself provides an excellent description of what a Dockerfile is:
“Docker can build images automatically by reading the instructions from a
Dockerfile. ADockerfileis a text document that contains all the commands a user could call on the command line to assemble an image.” — Docker
Designed & Developed by Jasper Anders