 |
Ten years in the past, we introduced the final availability of Amazon Aurora, a database that mixed the pace and availability of high-end industrial databases with the simplicity and cost-effectiveness of open supply databases.
As Jeff described it in its launch weblog publish: “With storage replicated each inside and throughout three Availability Zones, together with an replace mannequin pushed by quorum writes, Amazon Aurora is designed to ship excessive efficiency and 99.99% availability whereas simply and effectively scaling to as much as 64 TiB of storage.”
After we began creating Aurora over a decade in the past, we made a elementary architectural resolution that might change the database panorama without end: we decoupled storage from compute. This novel strategy enabled Aurora to ship the efficiency and availability of economic databases at one-tenth the fee.
This is likely one of the explanation why a whole bunch of 1000’s of AWS prospects select Aurora as their relational database.
At this time, I’m excited to ask you to affix us for a livestream occasion on August 21, 2025, to rejoice a decade of Aurora database innovation.
A short look again on the previous
All through the evolution of Aurora, we’ve centered on 4 core innovation themes: safety as our high precedence, scalability to satisfy rising workloads, predictable pricing for higher value administration, and multi-Area capabilities for world functions. Let me stroll you thru some key milestones within the Aurora journey.
We previewed Aurora at re:Invent 2014, and made it typically out there in July 2015. At launch, we introduced Aurora as “a brand new cost-effective MySQL-compatible database engine.”
In June 2016, we launched reader endpoints and cross-Area learn replicas, adopted by AWS Lambda integration and the power to load tables immediately from Amazon S3 in October. We added database cloning and export to Amazon S3 capabilities in June 2017 and full compatibility with PostgreSQL in October that 12 months.
The journey continued with the serverless preview in November 2017, which grew to become typically out there in August 2018. World Database launched in November 2018 for cross-Area catastrophe restoration. We launched blue/inexperienced deployments to simplify database updates, and optimized learn situations to enhance question efficiency.
In 2023, we added vector capabilities with pgvector for similarity search for Aurora PostgreSQL, and Aurora I/O-Optimized to offer predictable pricing with as much as 40 p.c value financial savings for I/O-intensive functions. We launched Aurora zero-ETL integration with Amazon Redshift which permits close to real-time analytics and ML utilizing Amazon Redshift on petabytes of transactional information from Aurora by eradicating the necessity so that you can construct and preserve complicated information pipelines that carry out extract, rework, and cargo (ETL) operations. This 12 months we added Aurora MySQL zero-ETL integration with Amazon Sagemaker, enabling close to real-time entry of your information within the lakehouse structure of SageMaker to run a broad vary of analytics.
In 2024, we made it as easy as only one click on to pick out Aurora PostgreSQL as a vector retailer for Amazon Bedrock Information Bases and launched Aurora PostgreSQL Limitless Database, a serverless horizontal scaling (sharding) functionality.
To simplify scaling for patrons, we additionally elevated the utmost storage to 128 TiB in September 2020, permitting many functions to function inside a single occasion. Final month, we’ve additional simplified scaling by doubling the utmost storage to 256 TiB, with no upfront provisioning required and pay-as-you-go pricing based mostly on precise storage used. This permits much more prospects to run their rising workloads with out the complexity of managing a number of situations whereas sustaining value effectivity.
Most just lately, at re:Invent 2024, we introduced Amazon Aurora DSQL, which grew to become typically out there in Could 2025. Aurora DSQL represents our newest innovation in distributed SQL databases, providing active-active excessive availability and multi-Area sturdy consistency. It’s the quickest serverless distributed SQL database for all the time out there functions, effortlessly scaling to satisfy any workload demand with zero infrastructure administration.
Aurora DSQL builds on our unique architectural ideas of separation of storage and compute, taking them additional with unbiased scaling of reads, writes, compute, and storage. It gives 99.99% single-Area and 99.999% multi-Area availability, with sturdy consistency throughout all Regional endpoints.
And in June, we launched Mannequin Context Protocol (MCP) servers for Aurora, so you possibly can combine your AI brokers together with your information sources and providers.
Let’s rejoice 10 years of innovation
By attending the August 21 livestream occasion, you’ll hear from Aurora technical leaders and founders, together with Swami Sivasubramanian, Ganapathy (G2) Krishnamoorthy, Yan Leshinsky, Grant McAlister, and Raman Mittal. You’ll study immediately from the architects who pioneered the separation of compute and storage in cloud databases, with technical insights into Aurora structure and scaling capabilities. You’ll additionally get a glimpse into the way forward for database know-how as Aurora engineers share their imaginative and prescient and focus on the complicated challenges they’re working to unravel on behalf of shoppers.
The occasion additionally presents sensible demonstrations that present you methods to implement key options. You’ll see methods to construct AI-powered functions utilizing pgvector, perceive value optimization with the brand new Aurora DSQL pricing mannequin, and discover ways to obtain multi-Area sturdy consistency for world functions.
The interactive format consists of Q&A alternatives with Aurora specialists, so that you’ll be capable to get your particular technical questions answered. You can too obtain AWS credit to check new Aurora capabilities.
If you happen to’re taken with agentic AI, you’ll significantly profit from the classes on MCP servers, Strands Brokers, and methods to combine Strands Brokers with Aurora DSQL, which display methods to safely combine AI capabilities together with your Aurora databases whereas sustaining management over database entry.
Whether or not you’re working mission-critical workloads or constructing new functions, these classes will provide help to perceive methods to use the newest Aurora options.
Register right now to safe your spot and be a part of this celebration of database innovation.
To the following decade of Aurora innovation!