Operational database cluster

You can use the Operational Database Data Hub cluster definition in CDP to create an operational database workload cluster that has Apache HBase, Apache Phoenix, and additional services that are required such as HDFS and ZooKeeper.

Based on the CDP Environment that you select, the operational database Data Hub cluster uses Amazon S3 or ‎Azure Data Lake Storage Gen2 as the storage layer for HBase. This means, that the HFiles are written to the object storage, but the Write Ahead Logs (WALs) are written to HDFS.

Data Hub makes use of cluster definitions and reusable cluster templates. The operational database Data Hub cluster includes cloud-provider specific cluster definitions and prescriptive cluster configurations called cluster templates.

The Operational Database with SQL cluster definition consists of Apache HBase, Apache Phoenix, and the underlying HDFS and Apache ZooKeeper services that it needs to operate. Apache Knox is used for the proxying of the different user interfaces.

Each cluster definition must reference a specific cluster template. And, you can also create and upload your custom cluster templates. For example, you can export a cluster template from an existing on-prem HDP cluster and register it in CDP to use it for creating Data Hub clusters.

Figure 1. Video: Operational Database with HBase on Data Hub in CDP