Optimize Docker Images with Dive tool!

Docker images are the building blocks of containers, as they contain the instructions required to create and run them. However, some images may be larger, more complex, or less efficient than others, depending on how they are built and what they contain.

Which can affect the performance, security, and maintainability of your applications. Therefore, it is crucial to understand and optimize Docker images.

For this, you can use a tool called dive. In this article, we will show you how to install and use dive, what features it offers, and how it can help you with your Docker image analysis.

What is Dive?

Dive is an open-source tool that allows you to explore a Docker/OCI image, layer by layer, its contents and discover ways to shrink its size and improve its efficiency.

Dive works in the following steps:

1. Fetch the image from a local or remote source, such as Docker hub or local Docker daemon using Docker API.
2. Analyzing image layers
   1. Layers are the incremental changes that are applied on top of each other to form the final image.
   2. Each layer corresponds to a command in the Dockerfile that was used to build the image.
3. Dive then displays interactive CLI with three sections:
   1. A list of layers on the left side, with their sizes and commands.
   2. A file tree of the selected layer on the right side, with the files and directories that were added, modified, or deleted in that layer.
   3. A status bar at the bottom shows the total image size, the potential wasted space, and the efficiency score.

It allows you to interact using keyboard shortcuts or mouse clicks.

The potential wasted space is the amount of space that could be saved by removing unnecessary or duplicated files from the image. The efficiency score is a metric that indicates how well the image is optimized, based on the ratio of wasted space to total image size.

How to Install Dive?

There are different ways to install dive, depending on your operating system. Let's take a look at some of them.

The latest release of Dive tool could be found in GitHub release page.

Linux

1. Ubuntu/Debian

Download latest deb release and install with apt command:

export DIVE_VERSION=$(curl -sL "https://api.github.com/repos/wagoodman/dive/releases/latest" | grep '"tag_name":' | sed -E 's/.*"v([^"]+)".*/\1/')
curl -OL https://github.com/wagoodman/dive/releases/download/v${DIVE_VERSION}/dive_${DIVE_VERSION}_linux_amd64.deb
sudo apt install ./dive_${DIVE_VERSION}_linux_amd64.deb

2. RHEL/CentOS

Download latest rpm release and install with rpm command:

export DIVE_VERSION=$(curl -sL "https://api.github.com/repos/wagoodman/dive/releases/latest" | grep '"tag_name":' | sed -E 's/.*"v([^"]+)".*/\1/')
curl -OL https://github.com/wagoodman/dive/releases/download/v${DIVE_VERSION}/dive_${DIVE_VERSION}_linux_amd64.rpm
rpm -i dive_${DIVE_VERSION}_linux_amd64.rpm

3. Arch Linux

You can install dive from the Arch User Repository (AUR) using pacman or yay directly as follows:

pacman -S dive

4. Nix/NixOS

   On NixOS:

   nix-env -iA nixos.dive

   Other than Nix OS

   nix-env -iA nixpkgs.dive

Mac

If you use Homebrew, you can install dive using brew package manager:

brew install dive

If you use Mac Ports, you can install dive using port:

sudo port install dive

Or, you can download the latest Darwin build from the GitHub releases page and run it directly.

Windows

You can download the latest Windows build from the GitHub releases page and run it directly.

Alternatively, you can use Docker to run dive as a container:

docker run --rm -it \
  -v /var/run/docker.sock:/var/run/docker.sock \
  wagoodman/dive:latest

The above command will mount your local Docker socket into the container, allowing dive to access your local images.

Verify dive installation:

How to Use Dive?

To use dive, you need to have a Docker image that you want to analyze. You can either use an existing image from a Docker registry or a local Docker daemon, or you can build a new image from a Dockerfile.

To analyze an existing image, simply run dive with the image tag, ID, or digest:

dive <your-image-tag>

To build and analyze a new image, you can build an image:

dive build -t <some-tag> <context>

For example, to build and analyze an image from the current directory using the Dockerfile in it, you can run:

dive build -t my-image .

This command will build the image using docker build and launch the interface.

For example, to analyze the official nginx image, you can run:

dive nginx:latest

It will fetch the image from the Docker Hub and launch the below interface

Once you are in the dive GUI, you can use the following keyboard shortcuts or mouse clicks to interact with it:

How to Optimize Your Docker Image with Dive?

To optimize your Docker image (remove unnecessary stuff) with dive, you can follow these steps:

1. Check its total size, wasted space, and efficiency score of the image.
2. Identify layers that contribute most to wasted space and see what files they contain.
3. Use the filter and sort options to narrow down your search.
4. Review the commands that created those layers and see if you can modify them to avoid adding unnecessary or duplicated files.
   For example, use multi-stage builds, combine multiple commands into one layer, remove temporary files or caches, exclude unwanted files with .dockerignore, use smaller base images, etc.
5. Rebuild your image with the modified commands and re-analyze it with dive.

See if the wasted space and efficiency score have improved.

Repeat steps 2 to 5 until the image is optimized enough to your satisfaction.

Conclusion

This is how you can use dive to analyze existing images or build new ones, and interact with them to improve the performance, security, and maintainability of your applications using containers.

We hope you enjoyed this article and learned something new about dive and Docker images. Thank you for reading!