Preparing Source Data

Scenarios

Before being imported to a database, data is stored on the related host. The server that stores the data to import is called the data server. In this case, you only need to check the communication between the data server and GaussDB cluster and record the data storage directory on the data server. Check whether the server has robust system resources such as memory, handles, and disk space, based on the load of the import job.

If the data has not been uploaded to the data server, perform the operations described in this section to upload it first.

Procedure

1. Log in to the data server.
2. Create a file storage directory /input_data.

   mkdir -p /input_data

3. Upload source data files to the created directory.

   GDS parallel import supports data only in CSV, TEXT, or FIXED format. Ensure that the source data file meets the format requirements.