Enabling customer choice, agility, and performance with SQL Server on Red Hat

April 28, 2020

Content type
Best Practices

Since the release of SQL Server 2017 on Red Hat Enterprise Linux, the industry-leading SQL Server database has been available on Red Hat Linux technologies. Microsoft and Red Hat continue to jointly engineer SQL-related offerings on Red Hat and provide enterprise experience to customers with co-located support for production workloads. Our engineering teams work side-by-side on activities including joint architecture reviews, functional testing, and performance testing to ensure the best experience and performance possible for our mutual customers. SQL Server on RHEL has shown tremendous performance, evidenced by leading 1 TB and 30 TB non-clustered TPC-H price-performance benchmarks.¹

In response to the ever-growing demand for various deployment models, you can run SQL Server on-premises with RHEL 8, in the cloud with Azure Virtual Machines, and in containers with Red Hat’s OpenShift Container Platform. Additionally, this week at Red Hat Summit we made exciting announcements related to Azure Arc hybrid capabilities. Azure Arc enables deployment of Azure services anywhere and extends Azure management to any infrastructure across on-premises, edge, and multiple clouds.

SQL Server on RHEL 8, including UBI

You can be assured that the latest SQL Server and Red Hat innovations work well together. With SQL Server 2019, you can leverage features such as Intelligent Query Processing, data virtualization, improved developer experiences, and much more deployed on the Red Hat 8 environment of your choice. You can deploy SQL Server 2019 on Azure, or in your data center with virtual machines. You can leverage orchestrators like OpenShift to run SQL Server in a fully supported stack by Microsoft and Red Hat by using UBI-based images for SQL Server containers.

Learn more about how you can get started with SQL Server 2019 on Red Hat Enterprise Linux 8.

SQL Server on RHEL 8 Azure Virtual Machines

For customers interested in running SQL Server as Infrastructure as a Service (IaaS), the Azure Marketplace offers pre-configured SQL Server 2019 on Red Hat Enterprise Linux 8 Azure Virtual Machines. The latest technology from both Red Hat and Microsoft in the image allows for quick yet highly scalable and performant deployments. These images have high availability Red Hat repository enabled, which allows one to configure Availability Groups on SQL Server with RHEL VM on Azure. To maximize cost savings, customers can bring their SQL Server licenses with Software Assurance to the cloud with Azure Hybrid Benefit and may consider Azure Reserved VM Instances.

SQL Server 2019 Big Data Clusters (upcoming)

SQL Server 2019 became generally available on Nov 1, 2019. Among the capabilities introduced with this release is Big Data Clusters, a new deployment pattern for SQL Server that adds Apache HDFS and Spark for big data storage and analytics, which will be supported on Red Hat OpenShift in an upcoming cumulative update release of SQL Server 2019. This deployment pattern runs entirely as Linux container images on Kubernetes enabling customers to run their analytical workloads at any scale, on an integrated platform designed to derive new intelligent insights out of data.

Red Hat’s OpenShift container platform (OCP) is a market leader in commercially supported Kubernetes distributions and is one of few supported reference platforms for running SQL Server in containers. Red Hat and SQL Server engineering teams are closely partnering to enable support for BDC on this platform. It is key for our enterprise customers to unblock BDC deployments on a fully supported stack, that follows best practices and guidance unique to OCP. We are working to enhance our deployment model and enable customers to be able to follow this guidance, so privileged containers are not required. In addition to non-privileged, containers should be running as a non-root user by default. Where elevation of privileges will be necessary to enable better process isolation within a container, we will clearly describe what and why we require certain security policies.