Compatible Computer Types in UNIX Environments

Big endian, little endian, and bi-endian refer to the way an operating system orders integer values. These values are ordered in bytes, and the way in which bytes are ordered is called byte ordering. The ordering of bytes depends on the operating environment on which the integers were produced.

On big endian platforms, the value 1 is stored in binary and is represented here in hexadecimal notation. One byte is stored as 01, two bytes as 00 01, and four bytes as 00 00 00 01. On little endian platforms, the value 1 is stored in one byte as 01 (the same as big endian), in two bytes as 01 00, and in four bytes as 01 00 00 00.

If an integer is negative, the "two's complement" representation is used. The high-order bit of the most significant byte of the integer will be set to on. For example, -2 would be represented in one, two, and four bytes on big endian platforms as FE, FF FE, and FF FF FF FE, respectively. On little endian platforms, the representation would be FE, FE FF, and FE FF FF FF. These representations result from the output of the integer binary value -2 expressed in hexadecimal notation.

Different computers store numeric binary data in different forms. Hewlett-Packard and IBM store data in big-endian format. Linux and Tru64 UNIX store data in little-endian format. Solaris SPARC stores data in big-endian format, and Solaris x64 stores data in little-endian format.

SAS files can be transported between compatible computer types using various methods including NFS, FTP, and CD. For two computer types to be compatible, they must have the following characteristics in common:

The tables in this section show the compatible computer types for Release 6.12 through SAS 9.3. After each table, a brief explanation is provided.

You can move a Release 6.12 SAS data set that was created on a 32-bit HP-UX host to a 32-bit AIX host using NFS, FTP, or CD. Because HP-UX and AIX are compatible computer types, you can use the V6 engine to read the HP-UX data set on the AIX host.

The same 32-bit HP-UX data set can be moved to a 32-bit Intel ABI host. However, because these computer types are incompatible, you cannot use the V6 engine to read the HP-UX data set.

You can use the V9 engine (SAS 9.3) to read V6 data sets, but you cannot write to them.

For information about reading Release 6 data sets, see Reading Version 6 Files.

You can move a Version 8 data set that was created on a 32-bit Solaris host to a 32-bit HP-UX host using methods such as NFS, FTP, or CD. Because this data set was created on compatible computer types, you will be able to read this data set in SAS.

The same 32-bit Solaris data set can be moved to a 64-bit HP-UX host. Because these computer types are incompatible, SAS will use Cross-Environment Data Access (CEDA) to read this data set. For more information, see Reading Version 8 or Later Files from Incompatible Computer Types.

In SAS 9.3, the Data Representation field of the PROC CONTENTS output shows the compatible computer types for a SAS file. The following is a display of the PROC CONTENTS output.

In this example, the Data Representation field shows the compatible computer types for this data set. You can use NFS, FTP, or CD to move any data set to any computer and SAS will load the data set. If one data set has the same data representation, then SAS can read the data set natively. Otherwise, SAS must use CEDA. For more information about CEDA, see Reading Version 8 or Later Files from Incompatible Computer Types.

In SAS 9.3 for the Solaris x64 platform, PROC CONTENTS lists the following compatible computer types in the Data Representation field of the PROC CONTENTS output: SOLARIS_X86_64, LINUX_X86_64, ALPHA_TRU64, and LINUX_IA64.

The following table lists the possible values for the Data Representation field for SAS 9.3, and the corresponding computer types.