Build Databricks-Connected ETL Processes in Google Data Fusion

Google Data Fusion allows users to perform self-service data integration to consolidate disparate data. Uploading the CData JDBC Driver for Databricks enables users to access live Databricks data from within their Google Data Fusion pipelines. While the CData JDBC Driver enables piping Databricks data to any data source natively supported in Google Data Fusion, this article walks through piping data from Databricks to Google BigQuery,

Upload the CData JDBC Driver for Databricks to Google Data Fusion

Upload the CData JDBC Driver for Databricks to your Google Data Fusion instance to work with live Databricks data. Due to the naming restrictions for JDBC drivers in Google Data Fusion, create a copy or rename the JAR file to match the following format driver-version.jar. For example: cdatadatabricks-2020.jar

1. Open your Google Data Fusion instance
2. Click the to add an entity and upload a driver
3. On the "Upload driver" tab, drag or browse to the renamed JAR file.
4. On the "Driver configuration" tab:
   • Name: Create a name for the driver (cdata.jdbc.databricks) and make note of the name
   • Class name: Set the JDBC class name: (cdata.jdbc.databricks.DatabricksDriver)
5. Click "Finish"

Connect to Databricks Data in Google Data Fusion

With the JDBC Driver uploaded, you are ready to work with live Databricks data in Google Data Fusion Pipelines.

1. Navigate to the Pipeline Studio to create a new Pipeline
2. From the "Source" options, click "Database" to add a source for the JDBC Driver
3. Click "Properties" on the Database source to edit the properties

   NOTE: To use the JDBC Driver in Google Data Fusion, you will need a license (full or trial) and a Runtime Key (RTK). For more information on obtaining this license (or a trial), contact our sales team.

   • Set the Label
   • Set Reference Name to a value for any future references (i.e.: cdata-databricks)
   • Set Plugin Type to "jdbc"
   • Set Connection String to the JDBC URL for Databricks. For example:
     
     jdbc:databricks:RTK=5246...;Server=127.0.0.1;Port=443;TransportMode=HTTP;HTTPPath=MyHTTPPath;UseSSL=True;User=MyUser;Password=MyPassword;

     To connect to a Databricks cluster, set the properties as described below.

     Note: The needed values can be found in your Databricks instance by navigating to Clusters, and selecting the desired cluster, and selecting the JDBC/ODBC tab under Advanced Options.

     • Server: Set to the Server Hostname of your Databricks cluster.
     • HTTPPath: Set to the HTTP Path of your Databricks cluster.
     • Token: Set to your personal access token (this value can be obtained by navigating to the User Settings page of your Databricks instance and selecting the Access Tokens tab).

     Built-in Connection String Designer

     For assistance in constructing the JDBC URL, use the connection string designer built into the Databricks JDBC Driver. Either double-click the JAR file or execute the jar file from the command-line.

     java -jar cdata.jdbc.databricks.jar

     Fill in the connection properties and copy the connection string to the clipboard.

   • Set Import Query to a SQL query that will extract the data you want from Databricks, i.e.:
     SELECT * FROM Customers
4. From the "Sink" tab, click to add a destination sink (we use Google BigQuery in this example)
5. Click "Properties" on the BigQuery sink to edit the properties
   • Set the Label
   • Set Reference Name to a value like databricks-bigquery
   • Set Project ID to a specific Google BigQuery Project ID (or leave as the default, "auto-detect")
   • Set Dataset to a specific Google BigQuery dataset
   • Set Table to the name of the table you wish to insert Databricks data into

With the Source and Sink configured, you are ready to pipe Databricks data into Google BigQuery. Save and deploy the pipeline. When you run the pipeline, Google Data Fusion will request live data from Databricks and import it into Google BigQuery.

While this is a simple pipeline, you can create more complex Databricks pipelines with transforms, analytics, conditions, and more. Download a free, 30-day trial of the CData JDBC Driver for Databricks and start working with your live Databricks data in Google Data Fusion today.