Operate big software
at scale on any cloud

Juju is an open source application modelling tool. Deploy, configure, scale and operate your software on public and private clouds.

Visit the docs to install Juju ›

New: Juju-as-a-Service is here

  • Hosted Juju controllers operate all the machines in your running models.
  • Deploy to all major public clouds — directly from your browser — or the command line.
  • Share your models and easily collaborate with other users.

Try the beta

Juju at your command


Featured solutions

Kubernetes MasterKubernetes the easy way

Deploy in minutes to all the major public clouds, private infrastructure or developer laptop.

  • We offer both a 5 node developer cluster and an 11 node production-grade cluster.
  • Juju actions enable your team to upgrade, pause and resume, snapshot and restore your nodes.
  • Validate your clusters to the core Kubernetes specification.
  • Scale the Kubernetes master and worker nodes independently.
View bundle

HadoopApache Processing MapReduce

Write applications which process vast amounts of data in-parallel on large clusters of machines.

  • Scale up from single servers to thousands of machines, each offering local computation and storage.
  • Deliver a highly-available service on top of a cluster of computers, detect and handle failures at the application layer.
View bundle

OpenstackOpenstack Base

Deploy the world’s most stable and manageable cloud on a single machine or start building a production cloud on a cluster.

  • A fully integrated and optimised combination of the latest releases of Ubuntu Server and OpenStack.
  • Provides dashboard, compute, network, block storage, object storage, identity and image services.
View bundle

Why use Juju?

Whether it involves deep learning, container orchestration, real-time big data or stream processing, big software needs operations to be open source and automated.

Juju is the best way to encapsulate all the ops knowledge required to automate the behaviour of your application.


Bundles are collections of charms that link services together, so you can deploy whole chunks of infrastructure in one go.

  • Percona Cluster
  • Keystone
  • Ceph
  • 13
OpenStack Base ›
  • Kubernetes Master
  • Kubeapi Load Balancer
  • Flannel
  • 3
Canonical Kubernetes ›
  • Rsyslog
  • Hadoop Slave
  • Ganglia
  • 8
Hadoop Spark ›

Charms contain scripts that simplify the deployment and management of a service.

Postgresql Postgresql › Prometheus Prometheus › Elasticsearch Elasticsearch ›


With Juju, relationships between linked charms are formed automatically.

Easily configure charms via the inspector.


Deploy and redeploy to all major public clouds or locally on your own hardware.

Monitor, maintain and scale on demand.

Join the movement


IRC channels on Freenode




Mailing lists

The Juju project


Ask the community

Find out more, get help or ask any questions at ask ubuntu.


Cloud MENA 2017

1-2 May 2017

Google Next 2017

8-10 March 2017
San Francisco

What people are saying about Juju

Spicule logo

Using the modelling ethos brought by Juju allows me to quickly run big data applications in a multitude of places.

Be it locally on my laptop, on bare metal or in the Cloud, Juju lets me reuse the same models and code without changing any aspects of my deployment.

Tom Barber
Spicule LTD
Ghent logo

Juju enables you to encapsulate each different part of your infrastructure and lets everything talk to each other.

So if you have a web server that’s managed by Chef and a database that’s deployed by a Docker container, you can have the web server talk to the database and the relations between these two very easily.

Merlijn Sebrechts
Ph.D. researcher
University Of Ghent
Epam logo

Juju acts as a thin layer on top of your infrastructure that allows all your operational code to talk to each other.

And because of this communication layer, a lot of the problems that conventional configuration management systems have just don’t exist anymore.

Konstantin Boudnik
Chief Technologist Bigdata, Open Source Fellow
EPAM Systems

Users and contributors

  • ubuntu
  • Deutsche Telekom
  • IBM
  • Intel
  • Microsoft
  • Cisco
  • China Telecom
  • NEC
  • HP
  • Yahoo
  • Verizon