saiku hadoop spark #2

  • By spicule
  • Latest version (#2)
  • Stable, Edge
  • 9 machines, 9 units

Add to new model

Saiku Enterprise Analytics with SQL

This bundle is a basic Saiku Enterprise Analytics deployment with Hadoop and Spark. It is designed to allow easy deployment of a basic OLAP analysis setup. The deployment of this bundle will deploy the following units:

  • 1 Tomcat with Saiku Analytics Charm as a subordinate to Tomcat
  • 1 HA Proxy
  • 1 HDFS Master
  • 1 Yarn Master
  • 1 Spark
  • 1 Thrift
  • 3 Compute Slaves

Usage

Deploy this bundle using juju:

juju quickstart u/spicule/saiku-hadoop-spark

or in Juju 2.0 or later:

juju deploy u/spicule/saiku-hadoop-spark

Interacting with the bundle

To make use of Saiku you need data. By default the MariaDB installation is empty, add some data to it!

To use Saiku you need to get a license, to install a trial license you can run the following:

 juju action do saikuanalytics-enterprise/0 gettriallicense

If your server hasn't had a trial license before, the charm should download and install one for you. If it fails, you can get a trial license from https://portal.meteorite.bi and install it with:

 juju scp "${MYDIR}/../var/license.lic" saiku/0:/tmp/license.lic
 juju action do saiku/0 addlicense path="/tmp/license.lic"