A way to persist data outside the container’s lifecycle.
They live on the host, not inside the container.
Useful for sharing data between containers or keeping data when containers are destroyed/recreated.

Types of Docker Volumes

Docker supports different types of volumes to handle data persistence:

Named Volumes

Managed by Docker.
Stored in a specific part of the host filesystem (like /var/lib/docker/volumes/).
Best for when you want Docker to handle storage location.

Example:

docker volume create my_volume
docker run -v my_volume:/app/data my_image

Automatically created by Docker when you mount a volume without specifying a name.
Harder to manage because they don’t have easy names.
Good for short-lived data.
Example:
```
docker run -v /app/data my_image
```
(Docker will create a random volume name.)

Example:

docker run -v /path/on/host:/path/in/container my_image

Data is stored in memory only (RAM), not on disk.
Super fast, but non-persistent (data disappears when the container stops).
Useful for sensitive information or temporary, high-speed data that you don’t want to be saved anywhere.
Works only on Linux hosts (natively).
Example:

```bash

docker run —tmpfs /app/cache my_image

```

(This mounts /app/cache as an in-memory filesystem.)

Type	Managed by Docker	Custom Path	Use Case
Named Volume	Yes	No	Persistent, portable data
Anonymous Volume	Yes	No	Temporary, unmanaged data
Bind Mount	No	Yes	Full control, dev workflows
tmpfs Mount	No	No	Non-persistent

Use named volumes for production.
Use bind mounts for local development.
Be careful with anonymous volumes — they can clutter your system if not cleaned.
Use tmpfs when you need speed and don’t want data saved.
Remember: tmpfs = in-memory only — no disk writes!