Learn how Azure DevOps supports, enhances, and accelerates the Analytics lifecycle in SAS Viya.
Learn How To
Use Azure Repos and Azure Pipelines from Azure DevOps.Create a self-hosted agent that will interact with SAS Viya on Azure. Create SAS Viya data, decision, model, scoring, and configuration pipelines.Who Should Attend
Technical resources working with or interested in the Azure cloud, DevOps concepts or SAS Viya automation and integration with Git and Azure.
Prerequisites
Before attending this course, you should be familiar with:;
Azure CloudSAS Viya fundamentals.;Familiarity with bash scripting, Python, Azure CLI (Command Line Interface), SAS Viya REST API (Application Programming Interfaces), SAS Viya CLI is also extremely helpful.SAS Products Covered
SAS Viya
Course Outline
Azure Pipelines Fundamentals (Microsoft hosted agent)
Create basic tasks and scripts.Work with parameters and variables.Structure the scripts in jobs and stages.Protect secrets in a pipeline. Work with an Azure Key VaultSelf Hosted AgentCreate a self-hosted agent to work with SAS Viya on Azure.Set-up a repository with necessary files.Understand Azure network communication. Open IP (Internet Protocol) and ports.Copy certificates.Register the self-hosted agent.Configuration pipelines: update packages, install needed Python version, install sas-viya cli and pyviyatools.The alternative to working with Virtual Machine self-hosted agents is using a Docker agent. Optional: Create an agent, Test an agent, Run a pipeline on two agents.SAS Viya Data PipelinesRun SAS programs in batch (in the pipeline and from a .sas file in Git).Add a repository in SAS Studio and trigger pipelines from there.Import and run a SAS Studio Flow.Create a complete data pipeline that loads a Cloud Analytic Services (CAS) table, imports a SAS Studio Flow, generates the SAS code from the flow, runs the code (flow), and stores the output in ADLS2 (Azure Data Lake Storage Gen2) blob storage.SAS Viya Data Pipelines 2Develop pipeline that calls a SAS Viya REST API and generates the SAS code from a SAS Studio Flow.Create a Python pipeline that runs the Python programs. The pipeline also creates and cleans output files and creates artifacts.Visual Studio Code: Work with SAS Viya directly from VS Code. Add the Azure Repo where programs are stored.SAS Viya Decision PipelinesCreate a Git publishing destination.Design a multi-stage decision pipeline: Import, Publish, and Deploy decisions.Optional: Register a SAS client and a secret. Score a published rule set using a REST API.Scoring the deployed decision using files stored in ADLS2 storage.SAS Viya Model PipelinesCreate an Azure publishing destination.Create a SAS Model Manager project and import models.Create a release pipeline. Every time a SAS model is published to Azure, it is deployed as container serving an Azure Web App. The new model is released into production without affecting the ongoing scoring.Automate the model import and register steps.
