Beyond the DATA Step: How to Filter and Alter CAS Tables Faster with PROC CASUTIL

When working in SAS Viya, the most common performance bottleneck is the unnecessary movement of data between the Compute Server and the CAS Server. While a traditional DATA step is versatile, it often requires "pulling" data out of the cloud to process it.

By using PROC CASUTIL, you unlock Server-Side Processing. Statements like PARTITION and ALTERTABLE execute entirely within the distributed CAS environment. This means your data never leaves the high-speed in-memory layer. Specifically, using ALTERTABLE with the keep= parameter is an "expert move" for memory management; it allows you to prune unnecessary columns from a table without the overhead of a full data rewrite. For massive datasets, this approach isn't just a best practice—it's the only way to ensure your cloud resources remain optimized and your execution times stay in the seconds rather than minutes.