Overview of Data Connections

The Data Management Server connects to data sources on databases through ODBC or through jobs that connect to the Federation Server. For more information about accessing data on or through a Federation Server, refer to the DataFlux Data Management Studio: User’s Guide and to the DataFlux Federation Server: Administrator’s Guide.
To add a data source using ODBC, use the ODBC Data Source Administrator provided with Windows, or use the dfdbconf command in UNIX and Linux.
You can configure the following data connections (also known as DSNs or data sources) for the jobs that you run on the DataFlux Data Management Server:
  • Domain-Enabled ODBC Connection - Enables you to create a connection that links a SAS MetaData Server domain to an ODBC Data Source Name (DSN). User credentials from the SAS Metadata Server are automatically applied when the user accesses the domain-enabled connection. This approach ensures that the appropriate credentials for that domain are applied to the access request.
  • Custom Connection - Enables you to create a custom connection string for non-ODBC connection types. These custom strings enable you to establish native connections from a SAS Federation Server to third-party databases, or to draw data from more than one type of data input.
  • SAS Data Set Connection - Enables you to create SAS data set connections.
In Windows, DSNs are stored in install-path\etc\dftkdsn.
In UNIX and Linux, DSNs are stored in install-path/etc/odbc.ini.
You can store ODBC credentials for data sources that require login credentials with the ODBC Credential Manager. With stored ODBC credentials, you can make connections to data sources without being prompted for login credentials. When a job is run, the saved user credentials are retrieved and used. The credentials are not stored within the job. The job references the connection by DSN only. In UNIX and Linux, credentials are stored in the directory /$HOME/.dfpower/dsn.
When you develop jobs and services in DataFlux Data Management Studio, use the Data Connections riser to set up and store login credentials for any Open Database Connectivity (ODBC) data source. The DataFlux Data Management Server can use these data sources directly if DataFlux Data Management Studio is installed on the same host as DataFlux Data Management Server.
Stored credentials do not have to be entered each time the job is run, and that information can be used by any DataFlux application. If you do not use stored credentials, then your job must authenticate through the SAS Metadata Server.
Use global variables within jobs and services to accept or retrieve data. Using global variables increases the flexibility and portability of DataFlux Data Management Studio jobs and services between data sources.
If you want to use ODB drivers other than those that are supplied, note that the DataFlux Data Management Server is compatible with most ODBC-compliant data sources. Also note that SAS provides limited support for drivers that are not supplied by SAS.
If you develop jobs that access a SAS Federation Server, then you can use JDBC drivers and other drivers that are written for native access to popular databases. To learn more about developing jobs that use Federation Server drivers, refer to the DataFlux Data Management Studio: User’s Guide and to the SAS Federation Server: Administrator’s Guide.
Last updated: June 16, 2017