Maintaining Processes

Example: Defining Data Transfer Process Properties to Move Data from Remote Host to Local Host

This example defines the Data Transfer process properties to move data from a remote host to the local host. A Data Transfer process is required for the Job because the input source and output table reside on different hosts. The properties specify that SAS/Warehouse Administrator will generate the necessary code to move the data. The Data Transfer process will copy the Customer input data, which is on a remote host, to an intermediate work table on the local host (the host where SAS/Warehouse Administrator is installed). The Mapping process then would map the columns from the intermediate work table to the target table, which is on the local host.

For an example about adding a Data Transfer process, see Example: Adding a Data Transfer Process to a Process Flow.

Note: The following explanation describes the metadata and methods used to achieve the desired results. It is assumed that the appropriate Data Warehouse Environment, Data Warehouse, Subject, data stores, Job, and Process Flow exist. [cautionend]

Defining Data Transfer Properties

In a Job Process Flow, a Data Transfer process can be added either below the output table or above the input source as appropriate. For this example, the Data Transfer process was added below the output table as shown in the following Process Flow:

[untitled graphic]

In the Job Process Flow, position the cursor on the Data Transfer process icon, click the right mouse button, and select Properties. The Data Transfer Process Attributes window displays.

General Tab	specifies a name, a description, an owner, and an administrator for the Data Transfer process.
Source Code Tab	specifies who supplies the source code for the process, which for this example is SAS/Warehouse Administrator. Note: When SAS/Warehouse Administrator generates the code, it uses SAS/CONNECT software and PROC UPLOAD or PROC DOWNLOAD to move the data.
Execution Tab	specifies the host on which to run the Data Transfer process, which for a Data Transfer is always the remote host.
Output Data Tab	specifies the name and location of the intermediate work table(s) produced by the Data Transfer process.
Options Tab	specifies options for generating the source code for the process. For this example, both options are selected, which are the default settings. The options available on this screen are Generate Access Code (ex. signon, libname) before step allows you to control whether SAS/Warehouse Administrator generates the code to access the preceding objects, such as sign on code, and the LIBNAME statement. If selected, SAS/Warehouse Administrator generates the code that is required to load or run the current object or process and the code that is required to access the objects or processes that precede the current one. It assumes that code has already been generated or is already in place to load or run the objects or processes that precede the current one. For example, when SAS/Warehouse Administrator generates code for a Data Transfer process, it generates the statements that are necessary to access the table that precedes the Data Transfer in the Job, but it will not generate any code that might be required to load that table. SAS/Warehouse Administrator assumes that the table has already been loaded. Set &SYSLAST macro variable at end of step allows you to control whether SAS/Warehouse Administrator generates code that sets the &SYSLAST macro variable to the name of the last table that was created or updated. Note: By default, both of the source code options are selected. If you specify that SAS/Warehouse Administrator generates code for the process (on the Source Code tab), then you should leave these source code options selected. If you are supplying the source code, you might need to select or deselect these options as needed.

Top of Page