Is it possible to define, in the syntax file(s) attached to a production job file from SPSS Statistics Production Facility, that the original data files can be imported from the server which SPSS Statistics Server was installed and the data processing will be done (interactively or as a background task)? If so, how do I declare, e.g. the file import in the GET DATA syntax command?

Let me put some context here. By default, the location of the imported/saved files points to the server, not to the computer which the client software is installed.

------------------------------
Estefano Souza
------------------------------

Hi @Estefano Souza

I believe you have just described the default behavior of IBM SPSS Statistics operating in distributed mode with Statistics Server.

So the answer to your question is "Yes".  When you are operating IBM SPSS Statistics client in distributed mode, the default behavior of the "File -> Import data" menus is to look on the Statistics Server for the data file to be imported.  Simply put,  build and test your command syntax file while operating IBM SPSS Statistics in distributed mode.  If the syntax runs correctly that way, then it will also run correctly from the Production Facility connected to the same remote Statistics Server.

I hope this helps!

Hello, David. Thanks for the answer.

One last question about it: if I create a production job file with SPSS Statistics Production Facility and run this file externally (e.g., via CMD ou Windows Task Manager) with login at the server where SPSS Statistics Server is installed, will the data file paths in the syntax file(s) also point to the server?

Hi @Estefano Souza

Yes.   File access to dat afiles is always from the perspective of the 'back-end' or 'server'.  When you are operating in distributed mode (Statistics client connected to Statistics Server) that 'backend' is the remote Statistics Server server.  When writing SPSS Statistics command syntax that access data files, always keep in mind that the data files will be accessed from the perspective of the 'server'. 

This is detailed in our online documentation:  Distributed Analysis Mode