Storage
Mobility
Protection
Analysis and control
TOPICShare
Azure SQL is Azure’s Platform as a Service (PaaS) version of Microsoft SQL Server. It has a complex service model with 18 possible combinations of deployment options, service tiers and compute models. In addition, there are two pricing models you can use to pay for Azure SQL—Database Transaction Units (DTU) and vCores.
In this post, we’ll provide a quick guide to the most common pricing options and briefly explain Azure SQL’s service and pricing, as part of our series of articles on Azure cost management. In addition, we’ll show how NetApp Cloud Volumes ONTAP can help save storage costs for Azure SQL deployment.
In this article, you will learn:
- Azure SQL Cost Examples: DTU Pricing
- Azure SQL Cost Examples: vCore Pricing
- Azure SQL Services: Deployment Models and Service Tiers
- Azure SQL Database Pricing Models
- Reducing Azure SQL Costs with Cloud Volumes ONTAP
Azure SQL Cost Examples: DTU Pricing
Database Transaction Units (DTU) is a way of measuring compute resources required to run your SQL database. Assuming you are already familiar with this model, pricing is covered below. If you are not, you can see our breakdown of the models below.
Below we show DTU pricing for Azure SQL using the General Purpose/Standard service tier, for a Single Database deployment, in the West US 2 region.
|
Instance Type |
DTUs |
Maximum Storage Allowed* |
Price for DTUs and Included Storage |
|
S0 |
10 |
250 GB |
$0.0202 |
|
S1 |
20 |
250 GB |
$0.0404 |
|
S2 |
50 |
250 GB |
$0.1009 |
|
S3 |
100 |
1 TB |
$0.2017 |
|
S4 |
200 |
1 TB |
$0.4033 |
|
S6 |
400 |
1 TB |
$0.8066 |
|
S7 |
800 |
1 TB |
$1.6130 |
|
S9 |
1,600 |
1 TB |
$3.2260 |
|
S12 |
3,000 |
1 TB |
$6.0488 |
* All options come with minimal Included Storage of 250GB
Additional data charges:
- If you use extra data storage beyond the included storage, the price is $0.17 per GB-month.
- If you need long-term retention of data using RA-GRS blob storage, the price is $0.05 per GB-month.
Azure SQL Cost Examples: vCore Pricing
Virtual cores (vCores) are a way of measuring your resource specs in a way that is comparable to on-premises server cores. This pricing model provides greater flexibility than the DTU model. When evaluating vCore pricing, there are two options you must consider — serverless compute and provisioned compute.
Serverless Compute Option
If you choose serverless compute, you can dynamically select between 0.5 and 16 vCores, with between 2.02 GB and 48 GB of memory.
The price is $0.5218 per vCore hour and $0.115 per GB-month.
Additional data charges:
- If you need point-in-time restore, you’ll pay $0.20 per GB-month for backup storage
- If you need long-term retention, you’ll pay $0.05 per GB-month
Provisioned Compute Option
In the provisioned compute option, you need to select a number of vCores and your capacity and prices are determined accordingly.
There is a regular vCore price including the Microsoft software license, and a lower price for Azure Hybrid Benefit, in case you transfer your existing SQL Server licenses to your vCores. In this scenario, the Azure Hybrid Benefit discount is 40%.
Below we show vCore pricing for Azure SQL using the General Purpose/Standard service tier, for a Single Database deployment, in the West US 2 region.
|
vCores |
Total Memory (GB) |
Price Per Hour Including License |
Price Per Hour with Azure Hybrid Benefit |
|
2 |
10.2 |
$0.5044 |
$0.3045 |
|
4 |
20.4 |
$1.0088 |
$0.6089 |
|
6 |
30.6 |
$1.5131 |
$0.9134 |
|
8 |
40.8 |
$2.0175 |
$1.2178 |
|
10 |
51 |
$2.5219 |
$1.5222 |
|
12 |
61.2 |
$3.0262 |
$1.8267 |
|
14 |
71.4 |
$3.5306 |
$2.1311 |
|
16 |
81.6 |
$4.0350 |
$2.4355 |
|
18 |
91.8 |
$4.5393 |
$2.7400 |
|
20 |
102 |
$5.0437 |
$3.0444 |
|
24 |
122.4 |
$6.0524 |
$3.6533 |
|
32 |
163.2 |
$8.0699 |
$4.8710 |
|
40 |
204 |
$10.0874 |
$6.0887 |
|
80 |
396 |
$20.1747 |
$12.1774 |
Additional storage charges:
- The vCore prices above only cover compute. You are charged separately for storage, using the Premium tier of the Azure Blob Storage service, at a cost of $0.115 per GB/month.
- If you need point-in-time recovery, you’ll pay $0.20 per GB-month.
- If you need long-term retention, you’ll pay $0.05 per GB-month.
To calculate Azure SQL costs for your specific scenario, see the official Azure pricing calculator. To better understand the related storage costs and compare different storage options, see NetApp’s storage-focused Azure calculator.
Azure SQL Services: Deployment Models and Service Tiers
To better understand Azure SQL pricing, it is helpful to understand the breakdown of deployment models and service tiers.
Azure SQL has three deployment models:
- Single Database—a fully managed database used by an organization.
- Elastic Pool—a group of Azure SQL Single Databases with a shared set of resources.
- Managed Instance—a fully managed Microsoft SQL Server Database Engine running in Azure, enabling easy migration of on-premises SQL Server databases.
There are three service tiers:
- General Purpose/Standard—for common workloads
- Business Critical/Premium—for high throughput OLTP applications requiring low latency and high resilience
- Hyperscale—for very large OLTP systems, performs auto-scaling of storage and compute
In addition, there are two compute models:
- Provisioned—Azure SQL provides Azure resources that run your database
- Serverless—the database is provisioned as a serverless component with auto-scaling compute and billing for use per second
So there are essentially 3 X 3 X 2 = 18 “flavors” of SQL database. Each flavor has its own pricing.
All pricing options are shown on the official pricing page, be sure to select the deployment model, service tier and computer tier at the top of the page to see the appropriate pricing:
Azure SQL Database Pricing Models
On top of the large variety of deployment options, there are also two pricing models you can use to pay for Azure SQL: The Database Transaction Unit (DTU) model and the vCore model.
DTU Pricing Model
Azure provides a DTU calculator you can use to measure how many DTUs your database requires. You can then select from one of several instance types or pool sizes (for the elastic pool deployment option). Each type offers a certain number of DTUs (representing the processing capacity) and a certain amount of storage.
vCore Model
A vCore model is suitable for larger databases. It gives you a certain number of virtual SQL servers running in Azure. You can select vCores with physical hardware characteristics of your choice (CPUs, number of cores, RAM, etc.) and a certain quantity of data storage, log storage, and attached backup storage.
An important benefit of the vCore model is that you can transfer your existing SQL Server licenses from your on-premise database servers to your Azure SQL vCores. This lets you use Azure Hybrid Benefit, a program in which Azure waives charges for Microsoft software as part of VM cost. Azure Hybrid Benefit lets you save up to 55% on vCore costs if you use existing licenses.
Switching Models
Microsoft documentation states that if your workloads require more than 300 DTUs, you should consider switching from a DTU to a vCore model. The rule of thumb for conversion is 100 DTUs = 1 vCore in the General Purpose tier, and 125 DTUs = vCore in Business Critical tier.
You can convert your SQL database at any time from a DTU to a vCore model, using an API or the Azure portal, with no downtime.
To see how Azure SQL pricing compares to other storage services, see our in-depth article about Azure storage pricing. Also learn about low-cost Azure storage options that can help you conserve costs.
Reducing Azure SQL Costs with Cloud Volumes ONTAP
NetApp Cloud Volumes ONTAP, the leading enterprise-grade storage management solution, delivers secure, proven storage management services on AWS, Azure and Google Cloud. Cloud Volumes ONTAP supports up to a capacity of 368TB, and supports various use cases such as file services, databases, DevOps or any other enterprise workload.
In particular, Cloud Volumes ONTAP provides storage efficiencies, including thin provisioning, data compression, and deduplication, reducing the storage footprint and costs for Microsoft SQL databases by up to 70%.
