There are three Azure SQL products with so many different deployment options, service tiers, and compute tiers that it can get quite confusing when choosing the right option for your workload. So, I thought I would write this blog to help out a bit.

Azure SQL is a cloud-based suite of database services designed to offer flexibility, scalability, and ease of management. It comprises three main products, each catering to different deployment needs and compatibility requirements. Within each of the products are various deployment options, service tiers, and compute tiers:

Below I expand on the options from the above diagram:

Azure SQL Products

1. SQL Server on a Virtual Machine (VM)

For organizations needing full control over their SQL Server environment, running SQL Server on an Azure VM is a great option. This Infrastructure-as-a-Service (IaaS) offering provides complete access to the operating system and database engine, enabling users to configure and manage SQL Server as they would on-premises. It is best suited for:

• Lift-and-shift migrations with minimal changes.
• Applications requiring full SQL Server features.
• Custom configurations and third-party integrations.
• See Provision SQL Server on Azure VM and VM size.

2. Azure SQL Database

Azure SQL Database is a fully managed, Platform-as-a-Service (PaaS) database solution that automates maintenance tasks like patching, backups, and scaling. It is ideal for modern cloud applications and is available in two deployment options:

Single Database

• A dedicated, isolated database with predictable performance.
• Best for applications that require resource guarantees at the database level.
• Supports service tiers: General Purpose, Business Critical, and Hyperscale.
• Can use the serverless compute tier (only in the General Purpose and Hyperscale tiers), allowing dynamic scaling of resources based on workload demand.
• See What is a single database in Azure SQL Database?

Elastic Pool

• Multiple databases share a set of resources, optimizing cost efficiency.
• Best for SaaS applications with variable workloads across multiple databases.
• Supports service tiers: General Purpose, Business Critical, and Hyperscale.
• Does not support the serverless compute tier.
• See Elastic pools help you manage and scale multiple databases in Azure SQL Database.

3. Azure SQL Managed Instance

Azure SQL Managed Instance is an instance-scoped deployment option that provides near 100% compatibility with SQL Server while delivering full PaaS benefits. This makes it ideal for organizations looking to modernize their database infrastructure with minimal friction. Key benefits include:

• Native support for SQL Server features like cross-database queries, linked servers, and SQL Agent.
• Built-in high availability, automated maintenance, and security.
• Supports service tiers: General Purpose and Business Critical.
• Does not support the serverless compute tier.
• See What is Azure SQL Managed Instance?

Service Tiers in Azure SQL

Azure SQL Database and Managed Instance offer three service tiers to cater to different workload needs:

1. General Purpose

• Balanced performance and cost-effective for most applications.
• Available in both Azure SQL Database (single and elastic pool) and Azure SQL Managed Instance.
• Supports the serverless compute tier (only for single database deployment).
• See vCore purchasing model – Azure SQL Database.

2. Business Critical

• Designed for applications requiring high transaction rates and low-latency I/O performance.
• Includes built-in high availability with multiple replicas.
• Available in Azure SQL Database (single and elastic pool) and Azure SQL Managed Instance.
• Does not support the serverless compute tier.
• See vCore purchasing model – Azure SQL Database.

3. Hyperscale

• Optimized for extremely large databases, supporting up to 128 TB.
• Provides rapid scaling of compute and storage independently.
• Available for Azure SQL Database (single database and elastic pool deployments).
• Supports the serverless compute tier for single database deployment.
• See Hyperscale service tier

Serverless Compute Tier

The serverless compute tier is a cost-effective option that allows automatic scaling of compute resources based on demand. It is available in the General Purpose and Hyperscale service tiers for Azure SQL Database (single database deployment).

Benefits of Serverless Compute Tier:

• Autoscaling of CPU and memory based on workload fluctuations.
• Automatic pausing of the database during inactivity, reducing costs.
• Best suited for intermittent, unpredictable workloads that do not require continuous availability.
• See Serverless compute tier for Azure SQL Database.

Choosing the right Azure SQL option

When selecting an Azure SQL offering, consider factors like workload requirements, compatibility, cost, and management overhead. Below is a quick reference guide:

Feature	SQL Server on VM	Azure SQL Database (Single)	Azure SQL Database (Elastic Pool)	Azure SQL Managed Instance
Fully Managed (PaaS)	No	Yes	Yes	Yes
Full SQL Server Compatibility	Yes	Partial	Partial	Nearly Full
Best for Lift-and-Shift	Yes	No	No	Yes
Best for SaaS Apps	No	No	Yes	No
Best for Modernization	No	Yes	Yes	Yes
Supports General Purpose	N/A	Yes	Yes	Yes
Supports Business Critical	N/A	Yes	Yes	Yes
Supports Hyperscale	N/A	Yes	Yes	No
Supports Serverless Compute	N/A	Yes (General Purpose, Hyperscale)	No	No

Conclusion

Azure SQL offers a diverse set of solutions to accommodate different business and technical requirements. Whether you need full control with SQL Server on a VM, a fully managed single database, a cost-effective elastic pool, or near full SQL Server compatibility with Managed Instance, Azure SQL provides the flexibility to optimize cost, performance, and scalability. Understanding these options and service tiers will help you make an informed decision tailored to your specific workload needs.

This diagram may help with choosing the right option for your use case: