In this article you can read,

• what strengths and weaknesses the MySQL databases of Azure, AWS and the Open Telekom Cloud have,
• to what extent these are suitable for e-commerce use cases,
• and where you can download the independent benchmark study for this case.

Databases from the cloud are convincing and have established themselves in the corporate environment. But which cloud offers the most powerful and cost-effective databases? benchANT has tested and compared MySQL offerings from AWS, Azure and the Open Telekom Cloud.

Databases are used in almost every IT architecture to manage and analyze data in a structured way. For decades, they have been one of the little-noticed but substantial components of IT. The market offers an almost unmanageable abundance of databases and database management systems.

Alongside the commercial products Oracle and Microsoft SQL Server, the open-source product MySQL is one of the most popular and widespread database systems in the world. The relational open-source system can cover a wide range of different application scenarios. Every hyperscaler, including the Open Telekom Cloud, offers a variant of it.

There are good reasons for using databases from the cloud. Above all, these include flexibility, cost and speed advantages. "As-a-service" databases unburden companies from license costs and allow deployment within minutes – including comprehensive database and infrastructure management. One of the first steps in a cloud transformation is often to replace databases (the so-called replatforming).

In a recent benchmark (March 2024), the benchmark company benchANT evaluated the performance and cost efficiency of MySQL databases on the Open Telekom Cloud (RDS MySQL) and compared them with corresponding offerings from Azure and AWS (AWS RDS for MySQL. MS Azure Database for MySQL - Flexible Server).

Two technical parameters were recorded for the evaluation of the databases: throughput and latency. The throughput indicates how many operations the database can process in a certain period of time. Operations include, for example, reading, writing, or deleting data in the database. Latency indicates the delays that occur during data processing. Both parameters affect the performance of the database – and, thus, the user experience with the respective application.

In order to achieve a realistic benchmark, benchANT drew on a common and widespread deployment scenario: the operation of an e-commerce application. In an e-commerce application, such as an online store, the focus is on transactional workloads (Online Transaction Processing - OLTP). An order entry system triggers search and write requests from various users on the database. This workload can be simulated and evaluated with the Transaction Processing Performance Council (TPC-C).

The result of the benchmark: all offerings have their strengths and weaknesses. However, the MySQL offering from the Open Telekom Cloud proved to be on a par with the two hyperscalers. The Open Telekom Cloud shone with the best throughput values and the best price/performance ratio for throughput (on par with AWS). The price/performance improves further when users take into consideration the included e-mail and phone support and the GDPR compliance guaranteed at Open Telekom Cloud.

Would you like to study the results in detail? Discover all the information about the benchmark in the benchmark study by benchANT. Similarly, a second benchmark for PostgreSQL was created at the same time. It was tested and compared for telecommunications and analytics workloads.