In a previous post, I discussed the public preview of Azure SQL Database Serverless. This is a newer product released from Microsoft for the Azure ecosystem. Moving to this new product is really easy to do and I thought that I’d show you how.
Moving the Database
From the screen shot below, you can see that I have a standard Azure SQL DB, Demo1, that is on the sqldbdemo-east server. In order to move this to serverless, first click on Configure located on the left hand menu under Settings.
On the next screen, we will need to make some decisions on how you want the database configured in this new compute tier. If the database is currently in the DTU price offering, you will need to move to the vCore-base pricing since that is the only pricing tier available.
Once we are in the vCore-based pricing, we can continue to configure the database.
The Serverless compute tier is only available through the General Purpose service tier so, if necessary, you will need to first click on that service tier to be able to see the Serverless Compute Tier. Once you are in the appropriate tier, just select the Serverless compute as shown in the image below.
Next, is the compute generation. For this particular tier, you are limited to the Gen5 generation which consists of up to 16 vCores and a maximum of 48GB of memory. You can also select the maximum number of vCores that you want. You have a limit of 16 cores and it is easily adjustable with the slider scale. In the image below, you will see that I have set this database to a maximum of 4 vCores.
You can also specify a minimum number of vCores to ensure that the database always has at least that number of vCores available. I’ve set it to a minimum of 1 vCores.
Next we can adjust the Auto-pause delay values. This is a valuable feature which allows you to automatically pause the database after a period of inactivity. You can set this delay up to 7 days and 23 hours which provides quite a bit of flexibility. Pausing the database will help save costs as when the database is paused you are only charge the cost of the storage and not the compute.
Next select the appropriate data max size. Notice that you can go up to 1 terabyte in size which gives you quite a bit of room to grow.
Finally, since it is still public preview, you must select the preview terms. Once you have done that, you can then click the Apply button and the scaling process will commence.
Once the scaling has been completed, the database is shown in SSMS like it was before.
Summary
The process to move from Azure SQL DB on a server to the new Serverless compute tier is quick and easy to accomplish. This new service tier might be a good fit for your particular workloads and I highly suggest you take a look at it. Keep in mind, however, it is still in public preview and as such I would not use it for production workloads until it is fully available. It is a great for for non-production environments, especially if those environments are not used 24×7, as you can help to save some costs by pausing them!
© 2019, John Morehouse. All rights reserved.
One Response
Keep in mind there are still some restrictions !!
The following features do not support autopausing.
That is, if any of the following features are used, then the database remains online regardless of the duration of database inactivity:
Geo-replication (active geo-replication and auto-failover groups).
Long-term backup retention (LTR).
The sync database used in SQL data sync.
ref: https://docs.microsoft.com/en-us/azure/sql-database/sql-database-serverless#autopausing-and-autoresuming