|
|
Ten years ago, we announced the general availability of the Amazon Aurora database, which combined the speed and availability of top commercial databases with the simplicity and cost efficiency of Open Source database.
As Jeff described in his start -up blog post: “The storage is replicated both across and across three availability zones, along with the update model controlled by the CVORA, Amazon Aurora is designed to provide high performance and 99.99% availability, while easily and efficiently scalled up to 64 TIB storage.”
When we began to develop Auror ten years ago, we made a basic architectural decision that would forever change the database landscape: we disconnected the storage from Compute. This new approach allowed Aurora to provide the performance and availability of commercial databases for one tenth of cost.
This is one of the reasons why hundreds of thousands of AWS customers choose Auror as their relational database.
Today I am excited to invite you to join us to a live event 21 August 2025 to celebrate the decade of aurora innovation.
A short view of back at the past
During the development of Aurora, we focused on four basic innovation topics: security as our highest priority, scalability to fulfill the growing load, predictable prices for better cost management and more regions for global applications. Let me go through several key milestones on the way of Aurora.
We looked at Aurora at Re: Invent 2014 and we made it generally available in July 2015. At the start we introduced Auror as a “new cost -effective database engine compatible with MySQL”.
In June 2016, we introduced the endpoints of the readers and the cross regions reading the replica, followed by the integration of the AWS Lambda and the ability to load the table directly from the Amazon S3 in October. In June 2017, we added cloning and export database and export to Amazon S3 and full compatibility with PostgreSQL in October of the same year.
The journey continued to preview without a server in November 2017, which was generally available in August 2018. The global database launched in November 2018 for recovery after a disaster in the area of the cross regions. We introduced a blue/green deployment to simplify database updates and optimized instances of reading to improve query performance.
In 2023, we added vector skills with a PGVector to search for similarity for Aurora PostgreSQL and Aurora I/O-optimized to provide predictable prices with up to 40 % of cost savings for I/O-intensive applications. We launched the Aurora Zero-ETL integration with Amazon Redshift, which allows almost analytics and ml in real time using Amazon Redshift on petabia data from Aurory by removing the need and maintaining complex data pipes that perform extract and load (ETL) operations. This year we added the integration of Aurora MySQL Zero-Etl with Amazon Sagemaker, which allowed almost access to real-time data in Lake architecture in Sagemaker to operate a wide range of analysts.
In 2024, we did it effortlessly, as just one click to choose Aurora PostgreSQL as a vector store for Amazon Bedrock knowledge base and launch Aurora PostgreSQL unlimited database, without horizontal scaling.
In order to simplify the scaling for customers, in September 2020 we also increased the maximum storage to 128 TIB, which allows many applications to work within a single instance. Last month, we further simplified scaling by doubled the maximum storage to 256 TIB without required any initial provision and prices paid by AS-You-go based on the actual storage used. This allows even more customers to operate their growing workload without the complexity of management of multiple instances while maintaining cost efficiency.
Last time on Re: Invent 2024, we announced Amazon Aurora DSQL, which was generally available in May 2025. Aurora DSQL is our latest innovation in distributed SQL databases, offering active active high availability and strong consistency. It is the fastest distributed SQL database without server for always available applications, effortlessly to suit any demand for zero infrastructure management.
Aurora DSQL builds on our original architectural principles of the storage and calculation department, with independent scaling of reading, writing, calculation and storage. It provides 99.99% of one-region and 9999% availability of multi-regions with strong consistency across all regional end points.
And in June we launched a model context (MCP) servers, so you can integrate your AI agents into your data sources and data services.
We will celebrate 10 years of innovation
By participating in the Livestream 21st August, you will hear from the technical leaders and founders of Aurora, including Swami Sivasubramanian, Ganapathy (G2) Krishnamoorty, Yan Leshinsky, Grant McALISTER and Raman Mittal. You will learn directly from architects who promoted the computing and storage department in cloud databases, with technical knowledge about Aurora architecture and scaling. You will also receive a insight into the future of database technology, as Aurora engineers share their vision and discuss the complex challenges that work to solve customers.
The event also offers practical demonstrations that show you how to implement key features. You will see how to create AI -driven applications using PGVector, understand optimizing the cost of the new Aurora DSQL price, and learn how to achieve a strong consistency for global applications for multiple regions.
The interactive format includes Q&A with Aurora experts, so you will be able to answer your specific technical questions. You can also receive AWS credits to test new Aurora skills.
If you are interested in AI AI, you will especially benefit from sessions on MCP servers, on Strads Agents, and how to integrate Strads with Aurora DSQL, showing how to safely integrate AI capabilities with your Aurora databases while maintaining database access.
Whether you run a workload of a critical mission or create new applications, these sessions will help you understand how to use the latest Aurora features.
Sign up today to secure your place and be part of this database innovation.
To the next decade Aurora Innovation!
– seb