Docker Image vs Container
What Is the Difference Between a Docker Image and a Container?
Docker image vs container comes down to one distinction: a Docker image is a read-only template, and a container is a running instance started from that image. In short, the image is the blueprint and the container is the working copy built from it.
An image is made of stacked layers. Each layer records one change, such as installing a package or copying application files. Layers are read-only and can be shared between images. When you build an image from a Dockerfile, most instructions create a new layer. If two images start from the same base operating system, the machine stores that base only once.
A container adds one thing on top of the image stack: a thin writable layer. When a container starts, Docker creates this layer so the process can write logs, temporary files and other changes. The image underneath never changes. You can start many containers from one image, and each one gets its own writable layer. For the wider tool, see what Docker is and how it works.
How Images and Containers Work Together in Practice
Consider a web application packaged with its runtime and dependencies. A developer builds one image from it. The team then starts three containers from that image to share incoming traffic. All three run identical code, but each keeps its own writable state.
The two have different lifecycles. An image is stored and can sit unused for months. It is pulled from a container registry when another machine needs it. A container is created, started, stopped and removed. Removing a container deletes its writable layer, so any file it wrote there is lost. This is why containers are treated as disposable and can be replaced without losing the image.
Data that must survive a container replacement belongs in a volume. A volume is storage that Docker manages outside the writable layer, so its contents remain available when a container is removed and a new one is started. Databases and uploaded files are typical examples.
Frequently asked questions
Can You Run a Docker Image Without Creating a Container?
Not as a running program. An image contains no running processes, so it does nothing until Docker creates a container from it. Starting that container launches the image's default command. You can create a container without starting it, but the application only runs once the container is started.
Does Deleting a Container Also Delete Its Image?
No. Deleting a container removes only its writable layer and its state. The image stays on the machine, and you can start new containers from it. Docker will not remove an image while a container still uses it, so you delete the image separately when it is no longer needed.
Where Should an Application Store Data So It Is Not Lost?
Store it in a volume or a bind mount rather than in the container's writable layer. Volumes are managed by Docker and suit databases. A bind mount links a folder on the host into the container. Both keep data outside the container, so it survives when the container is replaced.