IBM introduced the Data Set File System (aka dsfs) with z/OS 2.5, enhanced it in z/OS 3.1, and then ported the enhancements back to z/OS 2.5. It provides the z/OS UNIX System Services user with access to z/OS datasets as if those datasets were files in the UNIX filesystem. This allows any OMVS command to access any z/OS dataset (well mostly).

This post is about using sftp to download a PDF that is stored in a z/OS PDS to a workstation (Windows in this case) for use.

In this example I have:

1. Opened a command window
2. changed to my /temp directory
3. initiated sftp
4. changed to the /dsfs/bin/ filesystem with the subdirectory of cbt and the file967pds sub-sub-directory.
5. used the sftp dir command to verify the member is there
6. used the sftp get command to download the member into a file with a pdf suffix
7. then issued the start command for the file which then opens the pdf

Could anything be easier?