Getting Started

There are so many Azure Pipelines Extensions that it is hard to know where to start. This wiki will explain the basics of applying DevOps to your Data Warehouse.

The first steps you need to take are:

1. Set up your Development Desktop
2. Put your databases, cubes and SSIS packages under source code control by importing them into SQL Server Data Tools (SSDT)
   • Import Databases
     • Fixing Database Import Issues
   • Import SSAS Projects
   • Import SSIS Projects
   • Import SSRS Projects
3. Set up an Basic Build Pipeline which builds each of your components
   • Build Databases
   • Build Tabular Cubes
   • Build Multidimensional Cubes
   • Build SSIS Projects
   • Build SSRS Projects

When you have done all the above, you are ready to extend your Azure Pipeline by having the agent deploy the databases, cubes and SSIS packages to a build server, load test data and then run a set of automated tests against all the components. To do this, you will need an in-house self-hosted build server running the Azure Pipelines self-hosted agent for the reasons explained in our Build Server Setup guide.

Extend your pipeline to deploy your artifacts taking the following steps:

1. Set up a build server following our guidelines here: Build Server Setup
2. Deploy each of your components
   • Deploy Databases
   • Deploy Tabular Cubes
   • Deploy SSIS Projects
   • Deploy SSRS Projects
3. Check your components are working correctly by perform automated unit tests with one or more of the following tools:
   • SSDT Unit Tests
   • tSQLt Unit Tests

Take your pipeline to the next level by adding integration tests which prove all your components work together and give the results you expect.

1. Import Test Data
2. Run ETL Tasks
3. Run automated integration tests using tools such as
   • NBi Integration Tests
   • SpecFlow Integration Tests

Package your artifacts so they can be deployed using

• Azure Deployment