Oracle is making the Exadata Database Service on Exascale Infrastructure generally available on Oracle AI Database@AWS. At the same time, Oracle and AWS are extending their partnership agreement to accelerate migrations of Oracle databases to AWS. The service is now available in 22 AWS regions.
One year after the general availability of Oracle AI Database@AWS, both parties are taking the next step. The Exascale version of Exadata is now available in AWS environments, with a pay-as-you-go pricing model based on actual usage. Customers specify how much compute and storage they need.
The difference from the traditional Exadata setup lies in the shared infrastructure. Each database is distributed across a pool of storage servers, eliminating the need to reserve dedicated database and storage servers. According to Oracle, this also brings Exadata within reach of smaller databases and enables deployment across multiple regions for high availability and disaster recovery. Another new feature is instant thin cloning, which allows development and test copies to be created without duplicating storage.
From two regions to 22
The service offering is expanding rapidly. At general availability last year, Oracle Database@AWS was only available in US East (Northern Virginia) and US West (Oregon). By May of this year, the count had reached twenty regions; it now stands at 22 regions across Asia-Pacific, Europe, and North and South America.
“This momentum demonstrates that enterprises want a simpler, lower-cost path to modernize their most important Oracle workloads within AWS environments,” says Nathan Thomas, senior vice president of product management at Oracle.
The platform runs Oracle AI Database, which enables generative and agentic AI without moving or copying data. Through Oracle AI Vector Search, customers can search for meaning and context within their own business data.
The new long-term agreement also includes commitments to continued investment in Oracle AI Database@AWS and to accelerating customer migrations.