I assume you have a basic understanding of Docker. If not, you may refer to this previous article. I believe you understand the Docker importance in DevOps. Now behind this fantastic tool, there has to be an amazing, well-thought architecture. Isn’t it? But before I talk about that, let me showcase the previous and current virtualization systems.
Traditional vs. New-Generation Virtualization
Earlier, we used to create virtual machines, and each VM had an OS which took a lot of space and made it heavy. Now in docker container’s case, you have a single OS, and the resources are shared between the containers. Hence it is lightweight and boots in seconds.
Docker Architecture
Below is the simple diagram of a Docker architecture. Let me explain you the components of a docker architecture.
Docker Engine
It is the core part of the whole Docker system. Docker Engine is an application which follows client-server architecture. It is installed on the host machine. There are three components in the Docker Engine:
Server: It is the docker daemon called dockerd. It can create and manage docker images. Containers, networks, etc.Rest API: It is used to instruct docker daemon what to do.Command Line Interface (CLI): It is a client which is used to enter docker commands.
Docker Client
Docker users can interact with Docker through a client. When any docker commands runs, the client sends them to dockerd daemon, which carries them out. Docker API is used by Docker commands. Docker client can communicate with more than one daemon.
Docker Registries
It is the location where the Docker images are stored. It can be a public docker registry or a private docker registry. Docker Hub is the default place of docker images, its stores’ public registry. You can also create and run your own private registry. When you execute docker pull or docker run commands, the required docker image is pulled from the configured registry. When you execute docker push command, the docker image is stored on the configured registry.
Docker Objects
When you are working with Docker, you use images, containers, volumes, networks; all these are Docker objects.
Images
Docker images are read-only templates with instructions to create a docker container. Docker image can be pulled from a Docker hub and used as it is, or you can add additional instructions to the base image and create a new and modified docker image. You can create your own docker images also using a dockerfile. Create a dockerfile with all the instructions to create a container and run it; it will create your custom docker image. Docker image has a base layer which is read-only, and the top layer can be written. When you edit a dockerfile and rebuild it, only the modified part is rebuilt in the top layer.
Containers
After you run a docker image, it creates a docker container. All the applications and their environment run inside this container. You can use Docker API or CLI to start, stop, delete a docker container. Below is a sample command to run a ubuntu docker container:
Volumes
The persisting data generated by docker and used by Docker containers are stored in Volumes. They are completely managed by docker through docker CLI or Docker API. Volumes work on both Windows and Linux containers. Rather than persisting data in a container’s writable layer, it is always a good option to use volumes for it. Volume’s content exists outside the lifecycle of a container, so using volume does not increase the size of a container. You can use -v or –mount flag to start a container with a volume. In this sample command, you are using geekvolume volume with geekflare container.
Networks
Docker networking is a passage through which all the isolated container communicate. There are mainly five network drivers in docker: Conclusion I hope this gives you an idea about Docker architecture and its essential components. Navigate around Docker to learn more and if interested in getting hands-on training, then check out this Docker Mastery course.
