The Cloud Experience Everywhere

Azure SQL: How to move Databases to the Cloud [5 Steps]

Almost 90 percent of companies are now using cloud services on a daily basis. One particular motivator for this choice is the prospect of ending CAPEX commitments for good.

Once you’ve moved to Azure SQL, you want to remain there for the long haul, especially after migrating large databases over from your on-premises data centre. Fortunately, with the right planning and preparation you can move your SQL estate to Microsoft Azure and never worry about upgrading physical servers again.

statistic_id519639_share-of-companies-using-cloud-based-services-in-the-uk-from-2010-to-2017 (1)

88 percent of organisations are now using the cloud in their business

Want to know how it’s done? Use the steps outlined below to plan your move from on-premises SQL servers to Azure SQL with minimal downtime and risk.

Azure SQL Migration in 5 steps

Step 1: Assess your SQL Cloud Migration Needs

Understanding the infrastructure requirements of your SQL estate will make it easier to move to an Azure data platform. Firstly, you need to identify the simplest route to migration and the best way to move your SQL servers and databases in a cloud environment.

There are a few great tools for this purpose:

Each will give you a clearer idea of where additional configuration might be necessary, but for expert guidance consider partnering with a cloud data platform team.

Step 2: Choose your Preferred Azure SQL Cloud Destination (IaaS or PaaS)

You have two options for SQL migration depending on the size and function of your databases.

The first is PaaS (Platform-as-a-service) which allows you to hand control of the underlying plumbing to the Microsoft Azure team. Azure SQL databases are ideal for SaaS (software-as-a-service) applications and on-demand elasticity. Make sure you choose the right service tier to meet your database resource requirements.

Your second option is IaaS (Infrastructure-as-a-service) which enables you to directly transfer SQL servers to the Azure cloud in the form of virtual machines. This approach is perfect for legacy app migration where a high level of customisation is necessary.

Azure’s flexible infrastructure and tools means a mix of both solutions is often possible.

Step 3: Use the Right Migration Tools

Once you’ve settled on your final destination, it’s time to migrate. If you’ve opted for a PaaS solution, then you can export database copies using BACPAC files and fixing any errors with Transact-SQL scripts. Make sure you’ve selected the highest service tier your budget allows to optimise transfer performance (you can always scale down after migration to reduce costs).

For IaaS, you’ll need to spin up virtual machines to host your Azure SQL servers. To avoid downtime during migration, a lift and shift strategy is often the most effective method and is exactly how Microsoft moved their own HCP platform to Azure.

Step 4: Size your Azure SQL Databases for Maximum Performance

Now you’re set up on Azure, your main goal is database efficiency. To avoid in-house SQL server upgrades in the future, you need to ensure you size your databases correctly and configure them to scale within a certain budget.

With an Azure SQL database you can allocate resources for unpredictable periods of usage ahead of time. Elastic Database Pools give you the option group up to 100 databases and scale them in accordance with fluctuating workloads.

While IaaS is a little more work, this service is ideal for legacy databases that exceed Azure SQL database limits or conditions. Use this Microsoft guide to work out which of your applications better suit this model.

Final Step: Reduce your SQL Server Commitments

Whilst moving legacy databases to the cloud can feel like a daunting prospect - even Microsoftfaced this conundrum - with the right due diligence and planning, you can strike the perfect balance between on-premises and in cloud. This is ideal for organisations whose apps already exist in Azure, since moving data to the same service can help reduce latency issues.

Plus, if you’re using Azure SQL Databases, you’ll get the added bonus of outsourced OS management, giving your more time to work on the efficiency and user-experience of your applications.

To discover more, check out the work of our experienced cloud data platform team and start your journey to the SQL cloud today.

0 Kudos
About the Author


HPE Pointnext Services experts share their insights on the topics and technologies that matter most for your business.