Apache Whirr is a set of libraries for running cloud services. Whirr is currently in the Apache Incubator.
You can also use Whirr as a command line tool for deploying clusters.
Find out more about Whirr at our wiki.
You can use Whirr's CLI or APIs to get started with Whirr.
Have you got a suggestion for improving Whirr? It's easy to get involved.
Whirr uses jclouds for provisioning, so in principle it should support all the cloud providers that jclouds supports. The following table shows the cloud provider and service combinations that have been tested.
|Rackspace Cloud Servers||Yes||Yes||Yes||Yes||Yes||Yes|
For development and local testing we are also supporting the BYON (bring your own nodes) jclouds provider. Check the recipes folder for a configuration sample.
Each release is tested by running the integration tests on Ubuntu Server 10.04. All setup scripts should also work on Centos 5.x but we don't have a formal testing procedure in place right now.
Download a release of Whirr from a nearby mirror.
The code that would become Whirr started out in 2007 as some bash scripts in Apache Hadoop for running Hadoop clusters on EC2. Later the scripts were ported to Python for extra features (such as EBS support) and a wider range of cloud providers. These Python scripts are available today in Whirr under contrib/python.
In May 2010 the Apache Whirr Incubator project was started to give a home to the existing work that had been done, but also to create a Java version using jclouds as the cloud provisioning library. jclouds supports many providers and has a very rich API for running code on instances, so it provides a very solid foundation for building Whirr on.