Top Stories from the Microsoft DevOps Community – 2019.08.16

Sasha Rosenbaum

DevOps is making strides into the data realm. Whether we call it DataOps, MLOps, or simply CI/CD for Data, it is becoming easier to automate schema updates and data transformation processes. This week the community shared some excellent articles on the topic.

No interest in data? No worries! There is some content for the front-end developers as well!

Create Multi Stage YAML CI/CD pipeline for deploying database changes using Maven, Liquibase and Azure DevOps
Database schema changes have always been more difficult to automate than updates to code, and despite the industry efforts, most databases out there are still updated manually. In this post, Mohit Goyal is taking on the challenge of creating a database CI/CD pipeline using Liquibase and Maven in Azure YAML pipelines. Using a strategy like this can help you keep your schema changes in source control, and progressively roll them out to all of your environments!

Using DBT to Execute ELT Pipelines in Snowflake
Taking it a step further, John Aven is bringing DevOps into data transformation. John’s post describes using an open-source Data Build Tool (DBT) to create an Extract Load and Transform (ELT) pipeline with Azure YAML pipelines, using DBT to compile, test and run the data transformation “models”. The ability to test your changes can truly make a difference!

DevOps In Azure With Databricks And Data Factory
Since we’ve started on the Data path, I will also bring in a slightly older post by Alexandre Gattiker. The post features a detailed walkthrough of an Azure DevOps CI/CD pipeline for Azure Data Bricks and Azure Data Factory based Big Data application which predicts bike rentals using time and weather information.

Azure DevOps ReactJS Unified Build and Release
Bringing it back to the front-end development, let’s look at a YAML pipeline for a ReactJS application. In this post, Dara Oladapo is building and deploying a ReactJS application onto an Azure App Service on Linux using a Linux agent in Azure YAML Pipelines. Dara also shared his full YAML pipeline on GitHub here. Great work leveraging the new multi-stage CI/CD YAML pipelines!

Build your Jekyll site and Deploy it on GitHub Pages with an Azure DevOps pipeline
And for those who would like to avoid dealing with any type of database, there are always static websites! In this post, Xavier Geerinck walks us through his experience of moving a Jekyll website build and deploy pipeline from Travis CI to Azure DevOps. Xavier is working with a GitHub repo and YAML pipelines on an Ubuntu agent.

If you’ve written an article about Azure DevOps or find some great content about DevOps on Azure, please share it with the #AzureDevOps hashtag on Twitter!

1 comment

Discussion is closed. Login to edit/delete existing comments.

  • John Aven 0

    I would like like to thank my co-writer for building the ADO pipeline and writing the content on the pipeline, Prem Dubey. Excellent SRE and a joy to work with.

    Sasha, thank you for sharing our article!!!

Feedback usabilla icon