created by GATK_Team
on 2017-12-30
Dataproc is Google’s Spark cluster service, which you can use to run GATK tools that are Spark-enabled very quickly and efficiently. To use it, you need a Google login and billing account, as well as the `gcloud` command-line utility, ak.a. [Google Cloud SDK](https://cloud.google.com/sdk/).
There are two ways to create and control a Spark cluster on Dataproc: through a form in [Google’s web-based console](https://console.cloud.google.com), or directly through `gcloud`. Using the form is easier because it shows you all the possible options for configuring your cluster, but `gcloud` is faster when you already know what you want, and you can automate it through a script. See [this tutorial](https://software.broadinstitute.org/gatk/documentation/article?id=11093) for details.
Once you have created your cluster on Dataproc, all you need to do to run GATK Spark tools on it is add a few arguments to your GATK command-line to tell GATK where to find your cluster. Note that your data will need to be located in Google Cloud Storage (GCS), and your file paths will need to be prefixed `gs://` accordingly. And be sure to look into [jar caching](https://software.broadinstitute.org/gatk/documentation/article?id=11085) to speed up the process!
Updated on 2018-01-08