The loading of a Teradata or Oracle table into SAS Cloud Analytic Services (CAS) might take longer to run in SAS Viya 3.5 than in SAS Viya 3.4. The issue might occur when the following is true:

• You use a Teradata or Oracle CASLIB statement to access the table.
• You use the CASUTIL, CAS, or FEDSQL procedures to load the data into CAS.
• Your table has columns that are defined as VARCHAR where the defined length of the variable is large while the actual length of the data is small.

Possible workarounds for this issue include the following:

• Create a Teradata or Oracle view that CASTs the VARCHAR columns to a shorter length VARCHAR.
• Include the NCHARMULTIPLIER=1 option to your CASLIB statement.

Click the Hot Fix tab in this note for a link to instructions about accessing and applying the software update.

This hot fix adds a new option, SCANSTRINGCOLUMNS=, to the CASLIB statement to help with this issue. When set to TRUE, this option causes the code to execute a separate query using the MAX(LENGTH(column)) syntax to scan each VARCHAR for the exact max length. The data connector then defines the length to the exact space that is needed for each of those columns rather than the maximum length. This option is not supported with parallel loads that use DATATRANSFERMODE= "AUTO" or "PARALLEL" or in a LIBNAME statement with the TERADATA engine.

Operating System and Release Information

Viya on Linux: An update for this issue is available for SAS Viya 3.5. For instructions on how to access and apply software updates, see the Updating Your SAS Viya software section in the SAS Viya 3.5 for Linux Deployment Guide at