Amazon RDS vs Azure SQL Database – Which One is Better?

Santosh Karla | 2 min read

Public cloud has revamped the way organizations manage their database operations. There are numerous cloud service providers in the cloud domain but AWS and Azure are the leading players. Both Amazon AWS and Microsoft Azure are top options for deploying MS SQL Server databases among all the others. The market share in Cloud computing for AWS and Azure combined is around 53%.

In this article, we discuss the key differences of both the cloud services and compare their offerings, features, benefits and limitations; so that you can make an informed choice that best suits your business requirements.

Overview

With Azure, there are two implementations, single instance and managed instance. Azure SQL database enables your applications to run on the newest SQL Server 2019 with the latest features. While Amazon RDS comes in various versions, there are numerous editions of SQL server from 2012 to 2019 in a single dedicated instance. You are free to choose your own version and edition from express, standard, web and enterprise based on your requirements.

Both these models have their own benefits, but as far as flexibility goes, AWS is the best because some legacy applications cannot run on the new database engines. With Azure, if you want to use old versions of SQL Server, you have to implement the Azure Virtual Machines (VM). Both Azure and AWS are Platform as a Service (PaaS).

Connectivity

Both services use different architecture for connectivity. AWS Server provides direct connectivity in the form of AWS direct connect. It enables you a dedicated network connection to AWS, whereas Azure SQL databases can be deployed alongside customer databases on the same Server in isolation. Hence Azure databases require a gateway to determine which database or Server your application is trying to connect.

But there is a disadvantage with this gateway routing mechanism, which can have an unfavorable effect on your applications that require quick responses. On the bright side, the gateway approach ensures better management and scalability with this level of architecture. It also allows seamless migration of databases in case of any upgrades in the system.

Database Services

AWS RDS Azure SQL
Microsoft SQL Server Complete Guidance
Licence Mobility
License Mobility plus
Azure Hybrid Benefits
Auto Patching
Auto Backup
Oracle Database Complete Guidance Complete Guidance
Low LatencyOCI
MySQL Minimal Guidance Some Guidance
PostgreSQL No Guidance Minimal Guidance
Maria DB Minimal Guidance

Database Services

AWS RDS
Microsoft SQL Server Complete Guidance
Licence Mobility
Oracle Database Complete Guidance
MySQL Minimal Guidance
PostgreSQL No Guidance
Maria DB No Guidance
Azure SQL
Microsoft SQL Server License Mobility plus
Azure Hybrid Benefits
Auto Patching
Auto Backup
Oracle Database Complete Guidance
Low LatencyOCI
MySQL Some Guidance
PostgreSQL Minimal Guidance
Maria DB Minimal Guidance

Data Services

Data Services AWS RDS Azure SQL
Relational RDS Aurora SQL Database
DB for MySQL
DB for PostgreSQL
NoSQL – Key/Value DynamoDB
Simple DB
Cosmos DB
Table Storage
NoSQL Document DynamoDB Cosmos DB
NoSQL Graph Neptune Cosmos DB
Data Warehouse Redshift SQL Data Warehouse
Cache ElastiCache Redis Cache

Data Services

Data Services AWS RDS
Relational RDS Aurora
NoSQL – Key/Value DynamoDB
Simple DB
NoSQL Document DynamoDB
NoSQL Graph Neptune
Data Warehouse Redshift
Cache ElastiCache
Data Services AWS RDS
Relational SQL Database
DB for MySQL
DB for PostgreSQL

NoSQL – Key/Value Cosmos DB
Table Storage

NoSQL Document Cosmos DB
NoSQL Graph Cosmos DB
Data Warehouse SQL Data Warehouse
Cache Redis Cache

Licensing

Both Azure and AWS offer license inclusive fee models where you don’t have to buy extra licenses. In this model, you only pay for the amount of time you have been running the database for. However, Azure also has a bring your own license (BYOL) model for their managed instances. BOYL model can be beneficial for certain companies as it offers huge savings.

With Amazon, if you already have Microsoft software, then you have an option of BYOL to the AWS Cloud, and it is totally subjected to license terms by Microsoft.

Security

When it comes to security, Azure and Amazon are pretty secure -They both offer service-generated encryption certificates or allow their customers to provide their own by the Key Management Service for Amazon and Azure Key Vault for Azure.

The data, backups and snapshots, are all encrypted with the instances running on AWS RDS encryption, Data in transit is encrypted with SSL.

Both offer Active Directory integration, but with some limitations. As for their directories, AWS RDS works with Microsoft Active directory while Azure only works with Azure directory service.

Additionally, Azure SQL database provides advanced options like Data masking and Advanced Threat protection outside their database engine.

Scalability

AWS RDS SQL Server scales up to the limits of what the hardware provides. You can scale up by adding IOPS, Memory and CPU cores and threads to your existing server. This process is pretty effortless and requires less downtime. Additionally, Amazon provides special maintenance windows during this process to minimize the impact on users. With the above mentioned Hardware there are several options you can choose from, based on the version or editions you have. Currently, there are 27 instances you can choose from with multiple resource options. These Instances range from 1GB to 256 GB memory with a single processing core, up to 64 cores.

With Microsoft Azure service, it’s a bit different. Azure SQL database has two models: DTU and vCore.
The Database Transaction Unit (DTU) is a single instance model, while vCore unit applies to Managed instances. DTU models enable precision on the performance of your database and improve the overall performance with minimal impact on your service.

Meanwhile vCore comes with a concept of serverless database. With this concept, the service is automatically paused after a certain period of inactivity. Hence you only pay for storage. This option is perfect for development and testing processes.

Conclusion

Both Azure SQL and Amazon RDS database services strive to offer the best experience and excellent services to their customers. While Amazon RDS offers a server-based model with backward affinity to applications that run on earlier versions of SQL Server, Azure SQL provides a dynamic platform that runs on the newest versions of SQL Server.

From the Security point of view to Data protection, both these services are pretty robust and capable. So the choice between the two totally depends upon your IT environment and the Technologies you are currently using.

New call-to-action