This article will explain how to use Arm Treasure Data with Pentaho Data Integration by using our JDBC driver. By combining Treasure Data with Pentaho, users can scale their existing Pentaho Data Integration environment to handle huge volumes of data.

Prerequisites

• Basic knowledge of Treasure Data. The Quickstart Guide is a good place to start.

Download Pentaho Data Integration (Kettle)

You can download Pentaho Data Integration (Kettle) from the link below. Version 8.2  was tested for this article

• Pentaho Data Integration (Kettle)

Download JDBC Driver

You can download the driver from the link below. The driver is still in beta, so any feedback would be appreciated.

• JDBC Driver Download

Copy JDBC Driver Jar to Pentaho Data Integration

Before starting Pentaho Data Integration, please copy the Treasure Data JDBC driver to the libext/JDBC directory specified by Pentaho Data Integration.

$ cp td-jdbc-VERSION.jar \
  <pentaho-data-integration-install>/data-integration/libext/JDBC/

Create Treasure Data Database Connection

Connect Treasure Data to Pentaho Data Integration’s database connection and make a new transformation. Please follow the procedure below.

Create a New Transformation

1. Open the Pentaho DI application.

2. Select File > New > Transformation

Create New Database Connection

1. Navigate to Tools > Wizard > Create database connection.
   
2. Edit the details in the pop-up dialog.
   
   • Name the database connection
   • Select Generic Database for type of database
   • Select Native (JDBC) for type of database access
     
3. Specify the connection’s URL and name of the driver class. URL can be specified in any of the following ways:

• jdbc:td://api.treasuredata.com/<db_name>
• use jdbc:td://api.treasuredata.com/<db_name>;useSSL=True  if you want to enforce SSL
• You can choose the query engine Hive or Presto (default) by the parameter
• jdbc:td://api.treasuredata.com/sample_db;useSSL=true;type=hive
• jdbc:td://api.treasuredata.com/sample_db;useSSL=true;type=presto

1. Specify your username and password. Use your Treasure Data credentials for these fields (Your User Name is the email address used to register on Treasure Data). 
   
   
2. Select Test database connection.

Use Treasure Data Database as Table input

Specify Table Input

Select Table input from the transform’s Input context menu. Drag and drop onto the workspace as shown below.

Edit the Table Input

Right-click the Table input icon on the workspace. Select Edit step from the context menu. Configure your Table input as shown below.

We cover the query above in the JasperSoft iReport with JDBC Driver article.

Confirm the Table Input

To confirm the behavior of the Table input, write out your data on Treasure Data to JSON output as shown below.

Tips: How can I use Pentaho through a proxy?

Check ‘start-pentaho.bat’ or ‘start-pentaho.sh’, and add the following parameters to CATALINA_OPTS option:

CATALINA_OPTS="-Dhttp.proxyHost=<proxy address> -Dhttp.proxyPort=<proxy port>"