Import Data from Hadoop Gateway

You can import the following file types from a Hadoop Distributed File System (HDFS): .avro, .csv, .json, .orc, .parquet, .txt.

If you choose to import files that don't have extension, you will be prompted by the Confirm File Type dialog box to identify the file type.

1. Log in to Web with administrator privilege and open a specific project.
2. Choose Add Data > New Data.

3. In the Data Sources dialog, place the mouse cursor over the Hadoop option and click Browse Hadoop Files.

The Connect to Hadoop dialog opens.

4. Drag the files that you want to import from the left pane to the right pane.

5. Click Finish to import the selected data files.

   or

   Click Prepare Data to preview the selected data.

6. Click Aggregation to apply functions or filters to your data import. For more information, see Apply Aggregation and Filtering to Hadoop Data Imports.

7. Optionally, click Wrangle to perform data wrangling.

   All preview and wrangling operations may not be available due to configuration and software constraints.

   After you preview and wrangle the data, click Finish to import the data.

8. Click either Connect Live or Import as an In-Memory Dataset in the Data Access Mode dialog box.

The connectivity timeout between Hadoop Gateway and Intelligence Server is 20 minutes by default. To increase this timeout limit, create a file named QueryDSServerTimeout.ini and place it in your Intelligence Server directory. The only entry in this file will be the numeric value (in minutes) for your timeout limit. Placing a value of -1 in this file will set the timeout to unlimited.