Docker: image vs container

Docker images and containers are responsible for running applications in isolated environments and starting those environments, respectively. They are therefore two of the most important elements of the Docker virtualization platform and the Docker ecosystem. Learn what a Docker image and container are, and what they have in common and what they are different from each other.

What is a Docker container?

Docker containers are isolated execution units for applications and their dependencies. In this respect, they are very similar to virtual machines (abbreviated as  » VM « ), but there are some important differences. The main difference is that the degree of insulation Containers are slightly lower than VMs because no additional operating systems need to be virtualized. Instead, containers share the operating system and core of the host system. This has some advantages that make application development easier and faster, but also some disadvantages.

Containers have a much lower resource consumption than virtual machines, so they perform better than VMs and can be launched faster. In addition, they are much more portable than VMs, because only the application to be executed and its dependencies need to be transported. The low overhead Docker containers offer another advantage: multiple containers can be run simultaneously on the same system.

A disadvantage of this lower degree of insulation is that the containers have a less good security than VMs. Since all containers have access to the operating system kernel, an “infected” container could cause serious damage. For applications that require a higher level of security, it is therefore preferable to use a VM.

What is a Docker image?

A Docker image is a read-only model which is used by the Docker engine to create a container. So, Docker containers can be considered as executable instances Docker images. This template contains the dependencies of the application to be run, the necessary runtime environment, and the application code itself. As a container template, images are used to create consistent and reproducible execution environments for applications.

Docker images themselves are stored as binary files, but can also be stored in human-readable form. These text files are called Dockerfiles and contain all the instructions needed to restore the image. Both Docker images and Dockerfiles are portable and can therefore be transported very easily from one system to another. Thus, Docker images can be stored in Docker repositories like Docker Hub and downloaded from there. Running container instances can also be saved as image files using the command docker commit.

Docker comparison: image vs container

Docker images and containers are actually the same thing, but in two different forms. An image is a read-only template for a container, which contains all the files, dependencies, and runtimes needed to run an application, as well as the application itself. A container is a running, non-write-protected instance of that image file, just as a running program is an instance of the program files installed on the hard drive.

So, an image is a binary file stored in the host computer's file system and a container is a running process that resides in the system's RAM. Just as a Docker image is used to create a Docker container, a running container can also be saved as an image. This in turn can be used to replicate the exact same container.