Repolode
dundee

dundee/gdu

Fast terminal disk usage analyzer written in Go, with an interactive console UI, non-interactive output, and an optional web interface.

Open-source alternative to:WinDirStatDaisyDisk

Stars
6,188
Forks
254
Open issues
59
Last commit
22 hours ago

What is it?

Gdu scans a directory tree and shows which folders and files take up the most disk space. Its main interface is a console UI where you move through directories, sort by name or size, and delete or trash items. It is designed mainly for SSDs, where its parallel scanning makes the most difference, though it also works on HDDs with a smaller speed gain.

Beyond the interactive view, gdu can print summaries or the largest files in non-interactive mode, which suits scripts and pipes. It can export the full analysis as JSON, store results in SQLite or BadgerDB for later reloading, and import a saved JSON file. A browser-based web UI with a donut chart and treemap is also available via the --web flag.

Gdu supports options for ignoring paths by pattern or .gitignore rules, excluding file types, limiting scans to one filesystem, and showing apparent size instead of disk usage. Installation is a single binary from the releases page, and a Docker image is also provided.

Key features

  • 01Parallel scanning that is fastest on SSDs.
  • 02Interactive console UI with sorting, navigation, and delete or trash actions.
  • 03Non-interactive output with summaries, top-N largest files, and depth limits.
  • 04JSON export and import, plus SQLite or BadgerDB storage for saved analyses.
  • 05Browser-based web UI with donut chart and treemap views.
  • 06Path, pattern, .gitignore, and file type filters for scans.

Use cases

  • Finding which folders are consuming space on a full server or workstation disk.
  • Generating disk usage reports in scripts using non-interactive output.
  • Saving a scan as JSON or a database to review later without rescanning.
  • Browsing disk usage over SSH in a terminal session.

Quick start

  1. 1

    Download the Linux binary from the releases page

    curl -L https://github.com/dundee/gdu/releases/latest/download/gdu_linux_amd64.tgz | tar xz
  2. 2

    Make the binary executable

    chmod +x gdu_linux_amd64
  3. 3

    Move the binary into your PATH

    mv gdu_linux_amd64 /usr/bin/gdu
  4. 4

    Analyze the current directory

    gdu

Things to know

  • The documented quick install uses a Linux binary, and other platforms need the INSTALL.md instructions.
  • The Docker run command requires --privileged to access the host filesystem.

In collections

Topics

FAQ

What is gdu?
Fast terminal disk usage analyzer written in Go, with an interactive console UI, non-interactive output, and an optional web interface. Gdu scans a directory tree and shows which folders and files take up the most disk space. Its main interface is a console UI where you move through directories, sort by name or size, and delete or trash items. It is designed mainly for SSDs, where its parallel scanning makes the most difference, though it also works on HDDs with a smaller speed gain.
Is gdu free to use? What license does it have?
gdu is open source under the MIT license. Check the license terms before using it in commercial products.
Is gdu still maintained?
Actively maintained. The last commit to gdu was on Nov 6, 2026.
What can I use gdu for?
Finding which folders are consuming space on a full server or workstation disk. Generating disk usage reports in scripts using non-interactive output. Saving a scan as JSON or a database to review later without rescanning. Browsing disk usage over SSH in a terminal session.
How do I install gdu?
Download the Linux binary from the releases page: curl -L https://github.com/dundee/gdu/releases/latest/download/gdu_linux_amd64.tgz | tar xz Make the binary executable: chmod +x gdu_linux_amd64 Move the binary into your PATH: mv gdu_linux_amd64 /usr/bin/gdu Analyze the current directory: gdu
Where can I get gdu?
The source code and installation instructions are on GitHub: https://github.com/dundee/gdu

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects