SAS® Viya® Overview

Slides and Notes

Table of Contents 2

SAS® Viya® Overview Slides and Notes was developed by Stacey Syphus, Peter Styliadis, and Catherine
Truxillo. Instructional design, editing, and production support was provided by the Learning Design and
Development team.

SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of
SAS Institute Inc. in the USA and other countries. ® indicates USA registration.
Other brand and product names are trademarks of their respective companies.

SAS® Viya® Overview Slides and Notes

Copyright © 2024 SAS Institute Inc. Cary, NC, USA. All rights reserved. Printed in the United States of
America. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying, or otherwise, without the prior written
permission of the publisher, SAS Institute Inc.

Course code VYOV42, prepared date 02DEC2024. VYOV42_001

Table of Contents 3

Table of Contents
Table of Contents .......................................................................................................................................... 3
Course Overview ........................................................................................................................................... 4
Lesson 1: Introduction to the SAS® Viya® Platform ...................................................................................... 5
Lesson 2: SAS® Viya® Applications .............................................................................................................. 19
Discover Information Assets with SAS® Information Catalog ................................................................. 21
Manage Data with SAS® Data Explorer ................................................................................................... 23
Explore and Visualize with SAS® Visual Analytics ................................................................................... 26
Develop Code and Flows with SAS® Studio............................................................................................. 28
Build Models with SAS® Model Studio .................................................................................................... 32
Manage Models with SAS® Model Manager........................................................................................... 34
Share and Collaborate with SAS® Drive .................................................................................................. 36
Build Decisions with SAS Intelligent Decisioning .................................................................................... 38
Lesson 3: SAS® Viya® Programming Features ............................................................................................. 40
Accelerate Code with SAS® Cloud Analytic Services (CAS) ..................................................................... 41
SAS® Viya® and Open Source Integration ............................................................................................... 47
SAS® Viya® Workbench for Developers................................................................................................... 51
Course Overview 4

Course Overview

The purpose of this course is three-fold. First, we provide a big-picture overview of the SAS Viya platform
and the included tools and features. Second, we introduce you to each application so that you can
determine which applications will be most relevant for you or your organization. Third, the quick-start
follow-along demos allow you to get your hands in the applications and learn the initial steps to access
data and run a program, design a report, or build a model. As you fine-tune your focus in the AI and
analytics lifecycle, you can follow up with in-depth training on any of the topics in this course.
Lesson 1: Introduction to the SAS® Viya® Platform 5

Lesson 1: Introduction to the SAS® Viya® Platform

You've made a great choice to partner with SAS in your journey through the analytics life cycle. What do
we mean by the analytics life cycle?

It's the story of starting with real questions, and then using your data to make decisions based on
understandable, actionable insights. Questions like: Is this drug safe? How do our customers feel? What
is the financial risk, and more?
Lesson 1: Introduction to the SAS® Viya® Platform 6

The process takes us through three primary phases: manage data, develop models, and deploy insights.

First, the Manage Data phase provides an agile approach to data access, preparation, and governance. It
enables greater reliability, speed and collaboration in your efforts to operationalize data and analytic
workflows.
Lesson 1: Introduction to the SAS® Viya® Platform 7

Second, in the Develop Models phase, data scientists use a combination of techniques to build, optimize,
and validate predictive models. They use machine learning, natural language processing, forecasting,
optimization, and other techniques to answer real-world questions.

And third, the Deploy Insights phase focuses on automating, monitoring, and retraining AI models. The
goal is to validate and deploy models as quickly as possible, while ensuring quality results and enabling
transparent decision-making.
Lesson 1: Introduction to the SAS® Viya® Platform 8

Your role in the analytics life cycle is unique.

You might be focused on a particular step, such as manage data...

Lesson 1: Introduction to the SAS® Viya® Platform 9

...or you might oversee the entire process. Regardless of your role, the unified collection of applications
and programming languages on the SAS Viya platform enables you to use the subset of tools that you
need for your job and then seamlessly share and collaborate with others.

Again, you've made a great decision to partner with SAS in your analytics journey. Jump right in to see
how SAS Viya can unlock the value of your data.
Lesson 1: Introduction to the SAS® Viya® Platform 10

Let’s take a deeper look into what makes SAS Viya distinct. It's our most modern platform, designed to
optimize and accelerate the AI and Analytics lifecycle. It helps you get more done and make better
decisions – all faster than your competition.

How does Viya do that? Our software delivers three critical qualities: productivity, performance, and
trust. Let’s briefly touch on each one.
Lesson 1: Introduction to the SAS® Viya® Platform 11

SAS Viya automates the entire AI and analytics lifecycle – from data pipelines to model development, all
the way to monitoring model performance. We call this taking the computer science out of data science.
SAS Viya integrates with APIs, open data formats and many programming languages so organizations can
get more value out of their existing investments. And SAS Viya enables collaboration across business
roles and skillsets. Simply stated, SAS Viya is the unifying force that delivers unprecedented productivity
gains in the development of AI in your organization.

The second quality Viya provides is performance. Performance is delivering speed and scale so that you
can be more productive, while simultaneously minimizing expenses. Let me share a recent external study
that proves how SAS is delivering on speed, scale, and cost.
Lesson 1: Introduction to the SAS® Viya® Platform 12

We asked a 3rd party Research Firm, The Futurum Group, to independently compare SAS Viya with a
competitive, Cloud-based, commercial Data and AI platform, as well as open-source alternatives, like
H2O and sparkML.

They compared the performance by analyzing who is running and finishing computational AI
workloads/algorithms in the Cloud faster. They ran more than 1500 tests across both simple and complex
algorithms on 100 configurations and different data sizes. And the results are impressive!
Lesson 1: Introduction to the SAS® Viya® Platform 13

When it comes to speed, the study proved that Viya is, on average, 30 times faster than all the
alternatives.

And it gets even better when they compared SAS Viya to another commercial Data and AI platform. The
speed difference here, on average, is 49 times faster.
Lesson 1: Introduction to the SAS® Viya® Platform 14

And for complex models, Viya is up to 326 times faster.

So, what does this mean for you? It means your teams can take on more projects and deliver more
results for no more than you’re spending now. Or it could mean you run the same workloads you have
today, and lower your cloud operating costs by more than 86%.
Lesson 1: Introduction to the SAS® Viya® Platform 15

With any powerful technology, there are risks. This brings me to the final quality of our software – trust.

It’s important to understand that AI is rooted in mathematics that study the past to make decisions for
the present and the future. But if data from the past is biased or underrepresented, then AI will
perpetuate future decisions that are also biased and could have unintended consequences.
Lesson 1: Introduction to the SAS® Viya® Platform 16

We must ensure that decisions made from AI on behalf of people are explainable, transparent and fair.
At SAS, we call this responsible innovation. We believe that responsible innovation must come from the
creators of the AI technology. And we believe we have the most comprehensive, trustworthy AI platform
for data scientists and developers on the planet. Here is why...

Within these 6 core technology categories,

Lesson 1: Introduction to the SAS® Viya® Platform 17

SAS offers features to help you deliver trustworthy AI. From privacy preservation to bias mitigation to
natural language explanation, SAS Viya offers a full set of important capabilities that will help you
innovate responsibly. Given the rapid evolution of the market, we continue to revisit this landscape to
ensure we are delivering the capabilities our customers need.

Let's now get a little more technical and examine the basics of the SAS Viya architecture.
Lesson 1: Introduction to the SAS® Viya® Platform 18

With the SAS Viya platform, you can access a variety of data sources such as relational databases, the
cloud, folder paths, and Hadoop, as well as streaming data and more. The access is done with two
servers: the SAS Compute Server and SAS Cloud Analytic Services, or CAS. Let's discuss each of these at
high level.

The Compute Server on the SAS Viya platform enables you to execute traditional SAS code. The Compute
Server is similar to the SAS®9 workspace server and is sometimes referred to as the SAS programming
runtime environment. The Compute Server is used If you have SAS programming knowledge or legacy
SAS programs. You can still use the Compute Server to run your traditional SAS code with little to no
modifications. The Compute Server also includes some open-source integration with Python and R.

The platform also includes the CAS server. CAS is the high-performance, in-memory, massively parallel
processing server available in SAS Viya.

• CAS enables you to use a variety of programming languages like SAS, Python, R, Lua, or Java to
process your data in parallel.
• SAS Viya point-and-click applications use the CAS server for creating dashboards and performing
machine learning, text analytics, forecasting, and more.
• SAS Viya industry-specific solutions also use the CAS server's MPP environment (for example,
SAS Anti-Money Laundering, SAS Model Risk Manager, SAS Customer Intelligence 360, and
others).
• Lastly, you can create custom applications for your organization using the CAS server and REST
APIs.

These two servers in the SAS Viya platform provide a powerful analytic engine built for the entire
organization, no matter the job role, skill level, or programming language.
Lesson 2: SAS® Viya® Applications 19

Lesson 2: SAS® Viya® Applications

The core applications of SAS Viya take you through the AI and analytics life cycle. Each application is web-
based, so all you need is a browser.

SAS Information Catalog enables you to capture and enrich metadata for files, tables, and other
information assets. It can be used as the repository of metadata, including information on entities,
attributes, relationships, and domains. It enables you to discover and document how data fits together,
plus it provides a common vocabulary around data.

SAS Data Explorer displays views of data sources and provides you with access to additional data
management features, such as connecting to external data sources and loading tables into memory.
Lesson 2: SAS® Viya® Applications 20

SAS Visual Analytics helps you to visualize and discover relevant relationships in your data. You can
create and share interactive reports and dashboards, and use self-service analytics to quickly assess
probable outcomes for smarter, more data-driven decisions.

SAS Model Studio includes a suite of integrated tools to facilitate end-to-end data mining analysis. The
tools that appear in Model Studio are determined by your site's licensing agreement. Model Studio can
be used with SAS Visual Data Mining and Machine Learning, SAS Visual Forecasting, or SAS Visual Text
Analytics.

SAS Model Manager enables you to register, modify, track, score, publish, and report on analytical
models. It enables you to store models within folders or projects, develop and validate candidate
models, and assess candidate models for champion model selection – then publish and monitor
champion models.

SAS Studio is your tool for developing code and flows. You can access data files, libraries, and existing
programs – or write new ones with an enhanced program editor. You can also use predefined point-and-
click steps to generate code for you. These steps can be combined into flows to orchestrate data
preparation and analysis.

These six applications represent the core tools in the AI and analytics life cycle.

Depending on your SAS Viya license, you might have access to additional applications to perform other
important tasks.
Lesson 2: SAS® Viya® Applications 21

Discover Information Assets with SAS® Information Catalog

A key part of the Manage Data phase of the AI and analytic life cycle is to understand what data
resources you have available in your environment. We start with SAS Information Catalog, the
application designed to help you discover your information assets.

SAS Information Catalog provides the following key features for both data managers and consumers. It
creates and maintains an inventory for your information assets. It curates data sets, files, reports, flows,
and models. It enables you to search and explore asset metadata to discover the right resources for your
research. And it enables data administrators to review data usage from a single point of access.
Lesson 2: SAS® Viya® Applications 22
Lesson 2: SAS® Viya® Applications 23

Manage Data with SAS® Data Explorer

SAS Viya platform applications work with in-memory data on the SAS Cloud Analytic Services (or CAS)
server. We need the ability to access and manage our data sources in preparation for the remainder of
the analytics life cycle. SAS Data Explorer is the application that allows you to view and manage your
data sources, including in-memory data.

SAS Data Explorer provides the following key features. You can view and manage existing in-memory
tables and files. It provides the ability to access local files and load them into memory. And it guides you
through the process of creating connections to external data sources, including databases and Cloud
data.
Lesson 2: SAS® Viya® Applications 24

Before diving into the application, let's first define a caslib, which is key to understanding data access in
the SAS Viya platform. A caslib is the mechanism by which data is accessed on the CAS server. I like to
think that a caslib is to the CAS server what a library is to the SAS Compute Server. Caslibs can connect to
a variety of data sources in your organization like data in the cloud, databases, folder paths, and
streaming data.

However, caslibs work very differently, and it's important to understand how they are defined and used.

A caslib consists of three main parts.

First, there’s the data source portion. The data source portion contains connection information to a
physical data source for storage. Data might be stored on disk in a path with familiar formats such as SAS
data sets, CSV, Microsoft Excel, or parquet files. Or it may be stored in databases, the cloud, Hadoop, or
Lesson 2: SAS® Viya® Applications 25

other systems. The data source portion stores the connection and authorization information required for
interacting with those files using SAS Viya data connectors. We refer to data stored in the data source
portion of a caslib as files, even if the data source is a database table. Files in the data source area
cannot be processed by CAS directly.

The second part of a caslib is the in-memory portion. The in-memory portion provides a space to hold
tables on which operations can be performed. A data source file must first be loaded into memory. We
refer to data loaded into memory as tables, in-memory tables, or CAS tables. When files are loaded, in-
memory tables are available for processing, and they persist in memory. Because there's no connection
between the in-memory copy and the data source file, if you want to permanently keep changes made
to the CAS table during processing, the in-memory table should be explicitly written back to the data
source. If you are done with the CAS table, you can drop the table from memory.

Lastly, a caslib contains access controls to define permissions to that specific caslib.

Now let's see how SAS Data Explorer allows us to view and manage data in caslibs in our environment.
Lesson 2: SAS® Viya® Applications 26

Explore and Visualize with SAS® Visual Analytics

Integrated throughout the AI and analytics life cycle is the need to explore and visualize data. It is a
critical part of all phases: manage data, develop models, and deploy insights.

The SAS Viya application that makes this possible is SAS Visual Analytics. Regardless of your role, you will
likely use SAS Visual Analytics for a variety of purposes.
Lesson 2: SAS® Viya® Applications 27

It enables you to visualize and discover relevant relationships in your data, create and share interactive
reports and dashboards, and use self-service analytics to quickly assess probable outcomes.
Lesson 2: SAS® Viya® Applications 28

Develop Code and Flows with SAS® Studio

Developing code is integrated throughout the AI and analytics life cycle. We will primarily focus on using
code to manage and prepare data. However, it can be useful in all phases.

The SAS Viya application that is designed to access, develop, and execute code is SAS Studio.
Lesson 2: SAS® Viya® Applications 29

SAS Studio enables you to perform the following tasks:

• Access data files, libraries, and existing programs, or create new programs.
• Create and retrieve stored template code for common actions.
• Execute code on the SAS server and view results in your browser.
• Submit code for background processing or scheduled execution.

SAS Studio also enables you to build flows. In a flow, you construct a sequence of operations to prepare
your data. The actions in a flow are called steps. SAS provides a large collection of steps for a variety of
tasks, including executing SAS and Python programs, managing columns, joining tables, exporting files,
and generating reports and graphs.
Lesson 2: SAS® Viya® Applications 30

And you can go one step further and create your very own custom steps. A simple point-and-click visual
designer enables you to build an intuitive user interface that executes your customized SAS code behind
the scenes.

Although SAS Studio does provide some unique programming features, you have other options for
additional popular applications. SAS Enterprise Guide 8.4 or later offers the ability to connect to SAS®9
and SAS Viya. If you are a fan of Visual Studio Code, SAS offers a free extension with rich programming
features. Or if you prefer to use notebooks to code, you can also use JupyterLab.
Lesson 2: SAS® Viya® Applications 31
Lesson 2: SAS® Viya® Applications 32

Build Models with SAS® Model Studio

Developing models is the focal point of the analytics life cycle.

Building models is performed with SAS Model Studio. This SAS Viya application is an integrated visual
environment that provides a suite of analytic data mining tools to facilitate end-to-end data mining
analysis. These tools use the SAS Viya programming and cloud processing environments to create and
distribute analytic models, determine data mining champion models, score code, and deliver results.
Lesson 2: SAS® Viya® Applications 33

SAS Model Studio offers a user-friendly interface with a drag-and-drop approach to build various models,
such as regression or advanced machine learning algorithms. It supports model development and
comparison, including workflows for open-source models and custom SAS code, and allows users to
easily solve complex analytical problems with automated insights and pipelines. It provides a range of
pipeline templates and supports a variety of data preprocessing options. And Model Studio integrates
with SAS Model Manager for versioning, tracking, and deploying models. It also allows users to automate
API generation for retraining and scoring, and provides the ability to deploy models in database.
Lesson 2: SAS® Viya® Applications 34

Manage Models with SAS® Model Manager

After models have been built and optimized, it's time to put them to work. The Deploy Insights phase
allows you to automate, monitor and retrain your models.

SAS Model Manager is the tool for deploying and managing predictive models. It provides a centralized
repository for models and a suite of tools for managing the entire model development lifecycle.
Lesson 2: SAS® Viya® Applications 35

SAS Model Manager allows users to deploy models in multiple ways, including through REST APIs, SAS
Micro Analytic Service (MAS), and batch scoring. It includes tools for monitoring model performance and
detecting changes in model behavior over time, ensuring models remain accurate and reliable. It
facilitates collaboration on model development and sharing across teams. And it provides robust
governance capabilities, including version control, audit trails, and model validation.
Lesson 2: SAS® Viya® Applications 36

Share and Collaborate with SAS® Drive

SAS Drive is a hub for applications on the SAS Viya platform that enables you to easily view, organize, and
share your content from one place.

It provides convenient access to your most-used items, allows you to upload and download files in SAS
Content, enables easy sharing of assets with specific users or groups, and supports tagging and
commenting for all content.
Lesson 2: SAS® Viya® Applications 37
Lesson 2: SAS® Viya® Applications 38

Build Decisions with SAS Intelligent Decisioning

The ultimate goal of the analytics lifecycle is to use insights generated throughout the process to make
efficient and accurate business decisions.

SAS Intelligent Decisioning completes the analytics lifecycle, enabling you to operationalize analytics
models and combine them with business rules with a few clicks of a button, to drive automated
decisioning at scale. It supports customer-facing activities such as personalized marketing and next-best
action, plus decisions affecting customers, including credit services and fraud prevention.
Lesson 2: SAS® Viya® Applications 39

Automating decisions with SAS Intelligent Decisioning provides a streamlined mechanism for controlling
and monitoring the rules and processes used by your organization. You can integrate complex rule sets,
analytical and machine learning models, and custom code in a decision flow. It supports real-time or
batch execution of decisions, making it adaptable for various operational needs and enhancing the speed
and efficiency of decision-making processes. In addition, organizations that are highly regulated, such as
financial services, health care, and insurance, can more easily achieve compliance as a result of
repeatable, traceable decisions.
Lesson 3: SAS® Viya® Programming Features 40

Lesson 3: SAS® Viya® Programming Features

We've seen how SAS Viya guides you through all tasks in the analytics life cycle. However, remember
everything done in SAS Viya is driven by code. You may develop programs yourself, or code may be
generated for you through the point-and-click applications.

The Multi-Language Architecture is a key feature of SAS Viya, enabling you to seamlessly work across
multiple programming languages within the same analytics environment. This architecture supports a
combination of SAS, Python, R, SQL, REST APIs, and other languages, allowing for a flexible and
integrated workflow for data scientists, analysts, and developers.
Lesson 3: SAS® Viya® Programming Features 41

Accelerate Code with SAS® Cloud Analytic Services (CAS)

Recall that the SAS Viya platform includes two servers, the SAS Compute Server and SAS Cloud Analytics
Services, or CAS. The Compute Server is equivalent to the SAS®9 workspace server, and it executes your
traditional SAS code with little to no modifications.

The high-performance processing power of the SAS Viya platform is the CAS server. CAS provides the
run-time environment for data management and analytics with SAS. CAS is fast, reliable, and fault-
tolerant. In other words, it’s ideal for rapid processing of very large data for both data management and
analytics.

When you are running SAS®9 code in Viya, the Compute Server enables you to run your existing SAS
programs with few or no modifications, including DATA step code and all your familiar procedures for
visualization, analytics, modeling, and reporting.
Lesson 3: SAS® Viya® Programming Features 42

So, if you are a SAS®9 shop that is moving to Viya, don't be worried. You can still use all your SAS
knowledge!

If you are migrating SAS®9 code to SAS Viya, here are a few things to consider. First, you will likely need
to update paths or file references in your programs to point to the appropriate locations in your new
environment. Second, libraries referenced in your code must also be defined in SAS Viya. You can either
establish those libraries in your code, or your administrator can define permanent libraries. And third,
you can use the SAS Content Assessment tool to identify and evaluate the assets in your SAS®9
environment, including all SAS programs, to determine a migration path.

On the platform, we use the traditional SAS Compute Server to run SAS code and act as the client to the
CAS server. But what exactly is CAS?

The CAS server is a high-performance, massively parallel processing (or MPP) environment that is built to
process big data and your resource-intensive programs using modern analytics. It's also important to
know that all SAS Viya visual applications process data in this distributed computing environment.
Lesson 3: SAS® Viya® Programming Features 43

CAS works with in-memory data. The in-memory data sources can be processed in parallel to accelerate
execution speed. Data persists in memory for multiple reads until explicitly removed, reducing I/O (data
load from disk to memory) time.

Lastly, CAS uses sophisticated data connectors to access data across your organization. CAS can connect
to most of the same data sources as the Compute Server, or it can access other data sources. SAS Viya
data connectors were specifically designed for an MPP environment. There are many data connectors
available for popular data sources like Amazon Redshift, SingleStore, Snowflake, Spark, and more.

There are different ways to execute code in CAS. You can use CAS-enabled DATA and PROC steps to take
advantage of familiar SAS syntax. You can learn the native CAS language, CASL. Or you can leverage your
expertise in open-source languages, such as Python or R.
Lesson 3: SAS® Viya® Programming Features 44

There are several potential advantages of using the CAS server. Because data is loaded and held in-
memory, CAS avoids multiple reads of your data. CAS provides more processing power to accelerate
code execution. CAS offers additional analytical actions for reading in data and for machine learning,
deep learning, and image analysis. And CAS also prepares data for consumption in other SAS Viya
applications, including SAS Visual Analytics and Model Studio.

A caslib is the mechanism by which data is accessed on the CAS server. I like to think that a caslib is to
the CAS server what a library is to the SAS Compute Server. Caslibs can connect to a variety of data
sources in your organization like data in the cloud, databases, folder paths, and streaming data.

However, caslibs work very differently, and it's important to understand how they are defined and used.

A caslib consists of three main parts.

First, there’s the data source portion. The data source portion contains connection information to a
physical data source for storage. Data might be stored on disk in a path with familiar formats such as SAS
Lesson 3: SAS® Viya® Programming Features 45

data sets or CSV, Microsoft Excel, and parquet files, or it might be stored in databases, the cloud,
Hadoop, or other systems. The data source portion stores the connection and authorization information
required for interacting with those files using SAS Viya data connectors. Data stored in the data source
portion of a caslib is generally referred to as files, even if the data source is a database table. Files in the
data source area cannot be processed by CAS directly.

The second part of a caslib is the in-memory portion. The in-memory portion provides a space to hold
tables on which operations can be performed. A data source file must first be loaded into memory. Data
loaded into memory is generally referred to as tables, in-memory tables, or CAS tables. When files are
loaded, in-memory tables are available for processing, and they persist in memory. Because there's no
connection between the in-memory copy and the data source file, if you want to permanently keep
changes made to the CAS table during processing, the in-memory table should be explicitly written back
to the data source. If you are done with the CAS table, you can drop the table from memory.

Lastly, a caslib contains access controls to define permissions to that specific caslib.

So which programs should you consider executing in CAS? The answer is ”it depends.”

As you work in your SAS Viya environment and consider writing or modifying programs to run in CAS,
there are a few best practices to consider. CAS generally outperforms the Compute Server if you have
data sources larger than 50GB, programs that read data into memory multiple times, steps that take a
long time to run, PROCs that are computationally demanding, or DATA steps with complex logic.

You also want to use CAS to prepare data for use in other SAS Viya applications, such as SAS Visual
Analytics, or if you need a specific CAS action that does not have an equivalent procedure on the
Compute Server.
Lesson 3: SAS® Viya® Programming Features 46
Lesson 3: SAS® Viya® Programming Features 47

SAS® Viya® and Open Source Integration

There are so many open source language, package, and technology choices available today. At SAS, we
know our customers want to leverage their existing expertise, and use the best tool for each task.
Individually, these are powerful tools. Collectively, as you start to build analytics solutions, you quickly
find there are some important questions across this open source landscape.

Here are just a few of the questions and concerns that may arise when using open source solutions:

• Can we share the same platform while allowing data scientists to use their language of choice?
• How can we include non-programmers to make analytics accessible to all?
• How are models created, deployed, and governed? Are the results transparent and explainable?
• How do we access and connect to various data sources?
• How can we ensure our data science team is productive and the platform is performant?
Lesson 3: SAS® Viya® Programming Features 48

Fortunately, you don't have to choose between a commercially-supported analytics platform OR a

collection of open source tools. With SAS Viya, you can have it all in one integrated environment. Why
use SAS Viya with open source?

• It provides programming language choice. You can code in Python, R, or SAS.

• It democratizes analytics by allowing non-programmers to participate in the analytics life cycle
using powerful visual interfaces.
• It explains decisions with responsible and trustworthy AI.
• It reduces risk with end-to-end governance.
• It operationalizes analytics with simple model deployment.
• It allows you to access a wide variety of data directly with out-of-the-box data connectors.
• And it accelerates your analytics with a high-performance, reliable, supported architecture.

SAS Viya and open source integration works in both directions. While in SAS Viya applications, you can
leverage various open source languages. For example, in SAS Studio, you can execute Python code with
Lesson 3: SAS® Viya® Programming Features 49

the Python program editor or use PROC PYTHON in a SAS program. You can also execute R code using
PROC IML in a SAS program. In SAS Model Studio, you can build and score R or Python models using the
Open Source code node. And in SAS Visual Analytics, you can run a SAS Job that includes a call to R or
Python within the SAS program.

Using open source languages and technologies, you can also integrate with SAS Viya. SAS provides
several packages that allow you to use popular IDEs, such as Visual Studio Code, Jupyter Notebook or R
Studio. In these applications, you can create mixed workflows with Python, R and SAS. You can also
embed SAS Visual Analytics reports within a website using JavaScript to connect with SAS Viya using
APIs.

Visit developers.sas.com to access resources for integrating the SAS AI and analytics platform with open
source.
Lesson 3: SAS® Viya® Programming Features 50
Lesson 3: SAS® Viya® Programming Features 51

SAS® Viya® Workbench for Developers

SAS Viya Workbench is a lightweight, on-demand cloud development environment designed specifically
for programmers and modelers. Users can choose the amount of compute necessary for their project
and their preferred coding language. Currently, Workbench supports coding in SAS and Python through
Jupyter Notebooks and Visual Studio Code. Support for programming in R and through additional clients
is expected to be added in the future. SAS Viya Workbench is licensed separately, however it does
integrate with the enterprise SAS Viya platform.

One of the main benefits of SAS Viya Workbench is its centralized development for different users. SAS
and open-source developers can both code in the same environment and collaborate using GitHub for
distributed version control. This is all done in a lightweight implementation in which the users
themselves can limit their cloud costs by selecting only the amount of compute necessary and only using
it when they need it. With SAS Viya Workbench, users also have platform optionality, meaning they can
Lesson 3: SAS® Viya® Programming Features 52

tailor their environment to their specific needs and are not locked into a cloud or data provider. Finally,
Workbench can enhance performance through SAS's best-in-class analytics. Workbench currently
contains an extensive list of SAS procedures, covering a full range of data and analytics functionality, and
will continue to grow to cover all of SAS's AI and analytics technology. Open-source programmers can
achieve faster results by leveraging SAS analytics through APIs for Python with nearly identical syntax.
Currently, Workbench contains Python APIs that cover the most common statistical and machine
learning methods, with open-source capabilities continually expanding.

Users can deploy multiple Workbenches to accommodate a wide range of compute tasks. And these
Workbenches can then be connected to the enterprise SAS Viya environment to share, govern, and
deploy assets. By connecting to the enterprise analytics platform, content created in Workbench can be
serviced and managed across the organization.