This course focuses on learning DS2, a fourth-generation SAS language for advanced data manipulation. DS2 blends DATA step and SQL syntax with modern programming structures, and enables parallel processing in Base SAS as well as massively parallel processing environments such as Hadoop and SAS Viya Cloud Analytic Services (CAS). A brief introduction to SAS Viya is included.

Learn how to

• Identify the similarities and differences between the SAS DATA step and a DS2 data program.
• Convert a Base SAS DATA step to a DS2 data program.
• Create DS2 variable declarations, expressions, and methods for data conversion, manipulation, and conditional processing.
• Create user-defined packages and use predefined packages to store, share, and execute DS2 methods and extend the capabilities of the DS2 language.
• Create and execute DS2 threads for parallel processing.
• Leverage the SAS In-Database Code Accelerator to execute DS2 code outside of a SAS session.
• Execute DS2 code in CAS.

Who should attend

Experienced SAS software users who want to use modern programming techniques for complex manipulation of large data

Formats available	Standard Duration
Classroom:	2.0 days
Live Web Classroom:	4 half-day session(s)	System Requirements

This course is not appropriate for beginning SAS software users. Before attending this course, you should have several months of SAS programming experience or have taken the SAS® Programming 2: Data Manipulation Techniques course. A good understanding of ANSI SQL is also beneficial.

This course addresses SAS/ACCESS, Base SAS, SAS Viya software.

Introduction

• DS2 overview.
• Introduction to SAS Cloud Analytic Services (CAS).
• Course logistics.

Getting Started

• Hello, World.
• Basic DS2 syntax.

DATA Steps versus DS2 Data Programs

• Similarities to the DATA step.
• DS2 "missing" features.
• Converting DATA steps to DS2 data programs.

New Data Types and Syntax

• Program block structuring.
• Data types.
• Automatic data type conversion.
• Expressions.
• Selected functions.

Methods and Packages

• Methods.
• User-defined packages.
• Predefined packages.

Threaded Processing

• Threading on the Compute server.
• Threading on MPP platforms.