A batteries-included MapReduce cluster-in-a-can for scientists, researchers, and engineers.
Cirrus makes it easier to perform large-scale computation and data analysis tasks using the MapReduce framework. While MapReduce is widely used in industry, the high cost and complexity of setting up the MapReduce infrastructure make it impractical for many potential users. Cirrus Cluster is a python library and set of command line tools that automates the complex processes of provisioning, launching, configuring, and monitoring a custom MapR hadoop cluster. Cirrus aims to be a batteries-include tool to support a MapReduce workflow tailored for the needs of engineers and researchers who don’t have a full-time IT support staff to manage the cluster.
To get latest release...
pip install cirruscluster --use-mirrors
or to get current developement version...
git clone https://github.com/cirruscluster/cirruscluster.git; python setup.py install
Author: Kyle Heath (cirruscluster [at] gmail)