The Sahara project follows the same extensibility fashion as other OpenStack projects using the plugin terminology. Depending on which data processing framework a user or cloud operator would use, Sahara offers different sets of distributions by means of provisioning plugins. At the time of writing of this book, Sahara allows the provisioning of any data processing framework supporting one of the following vendors in OpenStack:
- Vanilla: Vanilla Apache Hadoop
- HDP: Hortonworks Data Platform
- MapR Distribution: MapR Filesystem (MapR-FS)
- Cloudera: Cloudera Hadoop
- Spark: Apache Spark
It is important to keep in mind that within every new OpenStack release, there are new, updated plugin versions. For the current setup, we keep using the Ocata release, supporting the following Sahara plugin versions:
# openstack dataprocessing plugin list
In the following...
