Reading SAS Data Sets From Previous Releases or From Other Hosts

Reading Version 6 Data Sets

Using the V6 and V6TAPE read-only engines, SAS can read Version 6 data sets that were created by compatible machine types. The following examples demonstrate how you can use the V6 engine.

• If you are running SAS 9.1 on Linux, you can use the V6 engine to read Version 6 data sets that were created with any Intel ABI release of SAS, such as SCO UNIX.

• If you are running SAS 9.1 on HP-UX, you can use the V6 engine to read Version 6 data sets that were created on HP-UX, Solaris, AIX, or IRIX.

For a list of the compatible machines types for V6, see "Compatible Machine Types for Release 6.12 through Release 8.2" on page 108. For more information about the compatibility of Version 6 files with SAS 9.1, see SAS Language Reference: Concepts .

Using CEDA to Read Data Sets

CEDA enables a SAS data set that was created in Version 7 or later in any directory-based operating environment (such as UNIX, Windows, and OpenVMS Alpha) to be read by a SAS session that is running in another directory-based environment. In SAS 9.1, if you try to access a data set that was created in a previous release, then SAS automatically uses CEDA to process the file. For example, if you are running SAS 9.1 on Linux, SAS will use CEDA to process a data set that was created in Version 8 on a 64-bit Solaris host. With CEDA, you have read and write access to these files; however, you will not be able to update the file.

For information about CEDA, see SAS Language Reference: Concepts .