Microsoft Softens Stance on SQL Server, Now Allowing License Portability to Amazon RDS
Table of Contents
A Rare Concession in the Cloud Wars
In a move that signals a pragmatic shift in enterprise cloud strategy, Microsoft is now allowing customers to apply existing SQL Server licenses toward usage on Amazon Relational Database Service (RDS). For years, the friction between Microsoft’s licensing regimes and Amazon Web Services (AWS) served as a significant barrier for enterprises attempting to move legacy databases into a fully managed environment without paying a ‘cloud tax.’
Previously, customers under Microsoft’s Software Assurance program could only bring their licenses to AWS via self-managed Amazon EC2 instances through the License Mobility program. While this allowed for cloud hosting, it didn’t offer the ‘hands-off’ experience of a Database-as-a-Service (DBaaS). To get a fully managed experience on RDS, users were effectively forced into a ‘License Included’ model, meaning they paid for the software a second time despite already owning the perpetual licenses.
The Mechanics of the Shift
The new “Bring Your Own License” (BYOL) framework for RDS for SQL Server allows customers using Enterprise or Standard Edition licenses to cover both the installation media and the licensing fees. According to AWS, the transition is designed to be a three-step administrative process rather than a technical overhaul.
First, customers must submit a License Mobility Verification Form to Microsoft to prove eligibility. Second, the SQL Server Release to Manufacturing (RMAN) media must be uploaded to an Amazon S3 bucket. Finally, within the Amazon RDS Console, users specify the SQL Server major and minor versions and point the service to the media file in S3. To maintain compliance and avoid the dreaded audit, AWS is integrating this workflow with the AWS License Manager, allowing admins to track license consumption in real-time.
The AI Imperative
While Microsoft has remained largely silent on the motivation behind this concession, the strategic benefit for Amazon is glaring. This isn’t actually a story about databases—it’s a story about data gravity and the race for AI dominance. By lowering the barrier for SQL Server migrations into RDS, AWS is effectively pulling vast amounts of proprietary corporate data closer to its own ecosystem of AI and analytics tools.
When data resides in a managed RDS environment, it is significantly easier to pipe into agentic AI applications and machine learning models. AWS explicitly noted that once operational data is in the cloud, teams can build AI applications that reason over business data without the “complex data pipelines” required when dealing with fragmented, self-managed EC2 setups.
Is SQL Server Losing Its Grip?
This move suggests a pivot in how Microsoft views its legacy software. For a long time, SQL Server was a primary lever used to pull customers toward Azure. However, the industry is shifting. According to the latest DB-Engines rankings, SQL Server currently holds the third position, but it has seen a steady decline in relative popularity over the last five years, with PostgreSQL rapidly gaining ground in the enterprise sector.
It appears Microsoft is less concerned with protecting the SQL Server moat and more focused on migrating users toward its modern data lake and analytics environment, Microsoft Fabric. By allowing BYOL on AWS, Microsoft may be acknowledging that forcing a binary choice between Azure and AWS is no longer a viable strategy for the modern multi-cloud enterprise.
Despite the slide in popularity compared to open-source alternatives, Microsoft’s database business remains a financial powerhouse. Gartner analyst Adam Ronthal recently observed that among the top vendors from 2011—including Oracle, IBM, and SAP—Microsoft is the only one that has consistently grown its market share over the last 15 years. If the goal is long-term ecosystem lock-in via AI and Fabric, allowing a few more SQL Server licenses to run on Amazon’s hardware might be a price Microsoft is more than willing to pay.
