emr-launcher 1.1.0

Launches an AWS Elastic MapReduce cluster using templated configuration files written in JSON. Meant to make deployments consistent and reproducable.

# EMR Launcher

Launches EMR clusters using config files for consistent run-time behavior when setting up a cluster.

## To Run

Starting a new cluster: ` PYTHONPATH=/path/to/emr_launcher python emr_launcher/launcher.py /path/to/config/<my_config>.json `

Adding steps to an existing cluster ` PYTHONPATH=/path/to/emr_launcher python emr_launcher/launcher.py config/<my_config>.json --job-id <job_id_of_existing_cluster> `

## Creating configs

Create a JSON file in the configs directory. Fill the config based on the parameters defined here: http://boto3.readthedocs.io/en/latest/reference/services/emr.html#EMR.Client.run_job_flow

or

build off the example config

## Using templating in configs

In any JSON config file function defined in the template_functions module can be used inside the config using jinja2 style templating. If you require a new function, add it to the template_functions module and it will be available to use in any config.