You can submit a Jaql job to InfoSphere BigInsights to transform and produce HDFS files for the Netezza. The JAQL_SUBMIT user-defined function (UDF) uses the ad hoc Jaql query application, so that application must first be deployed in InfoSphere BigInsights.

For larger data sets, users can encounter out-of-memory exceptions from the Jaql client. This article describes how to configure the heap size for ad hoc Jaql query applications:

  1. Deploy the ad hoc application.
  2. From the Files page in the BigInsights web console, go to user >applications, and select the application ID of the ad hoc Jaql application. Then select the workflow.xml file, and edit the contents in the contents pane of the console.filesystem
  3. Add the following property to the configuration section in the workflow.xml file:
    <property><name>oozie.launcher.mapreduce.map.java.opts</name><value>-Xmx2000m -Xms1000m</value></property>
  4. Your workflow.xml file will now look like this:XML
  5. Run the application.  The out of memory exception will no longer appear.

Join The Discussion

Your email address will not be published. Required fields are marked *