Accumulo on Google Storage. I use this to run Gaffer.

This is similar to my Accumulo -> {Zookeeper, Hadoop} stack, except that instead of Hadoop for HDFS, this bundles in the Google Storage connector with Accumulo, so that Accumulo can write directly to Google Storage.

Expects environment variables:

• HDFS_VOLUMES specifies the bucket location. Use a URI in the form: gs://MY-BUCKET/accumulo.
• NAMENODE_URI specifies the bucket location. Use a URI in the form: gs://MY-BUCKET/.
• PROJECT_ID name of my project.

Expects a private key of a service accounts with GS access at /key/private.json.