Linux on Microsoft Azure

Posted on July 5, 2021 by Adrian Wyssmann ‐ 2 min read

While starting to look at Azure, I am positively surprised on how Linux and open source (databases) are supported. Thus I want to mention a few words about it.


Linux infrastructure as a service

Linux is a first class citizen in Azure and there are different (Endorsed) Linux distributions - Microsoft contributes to the Linux kernel, ensuring that most newer Linux distributions are fully compatible with Azure.

Beside of virtual machines, the Linux-based platform as a service, Azure also offers other Linux-based services like Azure App Service, Azure Functions, Azure Container Instances, Azure Kubernetes Service and Azure Red Hat OpenShift, so you can easily deploy your Linux-based applications and workloads. I will explain these resources in more details, later.

Microsoft SQL Server on Linux

There are various reasons to choose a Microsoft SQL Server on Linux. Here a glance at the features

  • High Performance by use of Performance by using Hybrid Transactional Analytical Processing (HTAP)3, 4, which uses technologies like

  • High Security

    • Transparent Data Encryption (TDE) which encrypts data-at-rest when it is stored in database files
    • Always Encrypted ensures that only users who own data can view and process it, by encrypting it on client side.
    • Auditing tracks events that occur on the database engine and who executed those events
    • Row-level security controls access to specific rows within a table, based on who’s executing the query.
    • Dynamic data masking masks a portion of the data to protect sensitive data
    • Data Discovery and Classification identifies, labels, and reports on the sensitive data in your databases.
    • Vulnerability Assessment identifies vulnerabilities in your databases
  • High availability with different failover options like Always On availability groups, Always On failöver cluster instances

  • PolyBase enables your SQL Server instance to query data with T-SQL directly from SQL Server, Oracle, Teradata, MongoDB, Hadoop clusters, Cosmos DB without separately installing client connection software

  • Machine Learning Services allows you to run libraries against the data in your SQL Server databases to prepare your dataset, add features to it, train a model, evaluate the accuracy of the trained model

  • Graph support for storing and querying graph-based data

  • Full-text search enable users to run queries against text data that respect linguist rules

  • ETL workloads

Instead of VMs, you can also use containers as a lightweight alternative, to run your database.

Open Source DBs

Azure also offers databases such as MariaDB, MySQL and PostgreSQL either as IaaS or PaaS. For more details on planning the migration to Azure, checkout this. Keep in mind cannot use the Azure Database Migration Service yet, but using the SQL Server Migration Assistant.