Relational Database Offerings in Azure – Relational Databases in Azure

Relational Database Offerings in Azure

Until recently, most organizations hosted their database systems in on-premises datacenters that they owned or leased. They were responsible for applying updates to the database software and had to make sure that the hardware hosting the databases was properly maintained. Business continuity aspects such as database backup management, high availability (HA), and disaster recovery (DR) standards would need to be implemented to ensure minimal downtime in case of database corruption or server downtime. Scalability is also a concern, as database servers that outgrow compute allocated to them require someone to physically add compute to the server. All these items require additional hardware and levels of expertise from employees, thus increasing the total cost of ownership (TCO) for a database.

Cloud-based hosting has fundamentally shifted how organizations calculate TCO for their relational databases. Many operations that surround database upgrades or patching, business continuity, and scalability are handled by the cloud company. This allows organizations to shift their focus from maintaining hardware and managing business continuity concerns to being able to purely focus on the needs of the database users. Provisioning and scaling a database is also much easier as almost every requirement is preconfigured. Shortly put, databases can be easily deployed in Azure with the click of a button and scaled up and down with a slider (more on this later in the chapter).

Before getting into the different relational databases offerings in Azure, it’s important to understand the three types of cloud computing services. Having a foundational knowledge of how each of these are implemented is paramount to understanding the responsibilities and the TCO for hosting a database on Azure.

• Infrastructure as a Service (IaaS) offerings in Azure provide customers with the ability to create virtual infrastructure that mirrors an on-premises environment. IaaS offerings give organizations the ability to easily migrate their on-premises infrastructure to similar IaaS-based offerings in Azure without needing to completely redesign their applications using a cloud native approach. This is a typical first step for moving to the cloud as it allows organizations to offload the management of their hardware to Microsoft using a lift-and-shift strategy. While IaaS deployments allow organizations to no longer worry about maintaining the hardware powering an application, they will still need to manage maintenance at the operating system (OS) and application level. IaaS offerings include virtual machines that host services that would typically be hosted in a customer’s on-premises environment, such as SQL Server, and are connected via an Azure Virtual Network (VNet). These services can easily connect to an organization’s existing network infrastructure, allowing them to utilize a hybrid cloud strategy.
• Platform as a Service (PaaS) takes IaaS a step further by abstracting the OS and application software from the user. When deploying a PaaS offering, organizations can specify the resources they would like deployed, an initial size and compute tier depending on the intensity of the workload, what Azure region they would like them deployed to, and other optional or service-specific requirements. Azure will then provision the necessary resources to meet those specific requirements. Once deployed, all OS and software maintenance such as business continuity, upgrades, and patches are handled by Azure. This allows organizations to minimize the amount of effort required to maintain these services and instead focus on using them to build solutions that impact the business. Like IaaS offerings, PaaS offerings can also be interconnected via a VNet and connected to an organization’s existing on-premises network infrastructure. PaaS services include Azure SQL Database, Azure SQL Managed Instance (MI), and all the open-source relational database offerings that are hosted on Azure SQL.
• Software as a Service (SaaS) offerings represent the highest level of abstraction available to an organization hosting its infrastructure and applications on the cloud. Organizations simply purchase the number of licenses required for the service and then use it. Typical examples of SaaS offerings include Power BI Online and Office 365. None of the relational database offerings discussed in this chapter are SaaS offerings.