Oracle Database@AWS now supports Oracle Autonomous AI Database Serverless
Run fully managed Oracle Autonomous Database on AWS with zero infrastructure setup, serverless scaling, and native AWS service integrations.
View original announcement →Visual Summary
graph TD
A{{Oracle ADB-S on AWS}}:::announced
B((AWS Console/CLI)):::external
C([Auto-Scaling Compute]):::feature
D([4 AI Workload Types]):::feature
E(Amazon S3):::storage
F(AWS KMS):::compute
G(Amazon CloudWatch):::compute
H(Amazon EventBridge):::compute
I([Autonomous Data Guard]):::feature
B ==>|"provisions"| A
A -->|"backups"| E
A -->|"encryption"| F
A -->|"monitoring"| G
A -.->|"events"| H
A -->|"scales"| C
A -->|"supports"| D
A -->|"HA/DR"| I
classDef announced fill:#ff9900,stroke:#ec7211,color:#fff,font-weight:bold
classDef compute fill:#e3f2fd,stroke:#1565c0,color:#1565c0
classDef storage fill:#e8f5e9,stroke:#2e7d32,color:#2e7d32
classDef feature fill:#fff3e0,stroke:#e65100,color:#e65100
classDef external fill:#f5f5f5,stroke:#616161,color:#616161
What's New
Oracle Database@AWS has expanded to support Oracle Autonomous AI Database Serverless (ADB-S), a fully managed Oracle database service running on Exadata infrastructure that eliminates the need to provision dedicated hardware or VM clusters. Users can now spin up an Oracle Autonomous AI Database directly from the AWS Management Console, CLI, or APIs, with automated patching, tuning, and independent compute/storage scaling handled by Oracle. The service is available via public and private offers on AWS Marketplace, supporting both Bring Your Own License (BYOL) and License Included models.
How It Works
- Serverless provisioning: ADB-S runs on Oracle Exadata infrastructure managed by Oracle Cloud Infrastructure (OCI) inside AWS data centers, but users provision databases directly from AWS tooling without creating Exadata infrastructure or VM clusters.
- Four workload types: Supports AI Transaction Processing (OLTP), AI Lakehouse (analytics/data warehousing), AI JSON Database (document workloads), and Oracle APEX (low-code application development), each optimized for its respective pattern.
- Independent auto-scaling: Compute (measured in ECPUs) and storage scale independently and automatically based on real-time workload demand, with users paying only for what they consume in a serverless model.
- High availability and DR: Autonomous Data Guard provides built-in HA and disaster recovery, with automated backups written to Amazon S3 and cross-Region DR support for resilience against regional failures.
- AWS service integrations: Encrypts data using AWS KMS customer-managed keys, streams metrics and logs to Amazon CloudWatch for observability, and publishes lifecycle events to Amazon EventBridge for automation and alerting.
- Unified billing and access: Purchased through AWS Marketplace (public or private offer), the spend counts toward AWS committed spend (e.g., EDP) and qualifies for Oracle Support Rewards, with a single consolidated invoice.
Why It's Important
- Eliminates infrastructure overhead: Organizations no longer need to size, provision, or manage dedicated Exadata clusters to run Oracle Autonomous Database on AWS, dramatically lowering the operational barrier for Oracle workloads in the cloud.
- Accelerates Oracle-to-AWS migration: Enterprises running Oracle on-premises can migrate to a fully managed, serverless Oracle database without application changes, preserving full Oracle feature compatibility and performance characteristics.
- Bridges Oracle and AWS ecosystems: Deep integrations with AWS KMS, CloudWatch, and EventBridge mean Oracle databases can participate natively in existing AWS security, observability, and event-driven automation pipelines.
- Cost model alignment: The serverless, consumption-based pricing model aligns database costs directly with workload activity, making it financially viable for variable or unpredictable workloads that would over-provision on dedicated infrastructure.
- Supports AI-native workloads: The "AI" branding across all four workload types signals Oracle's positioning of ADB-S as a platform for AI Vector Search and generative AI application development, complementing AWS's own AI/ML services.
- Unified commercial relationship: A single AWS Marketplace invoice, AWS commitment eligibility, and Oracle Support Rewards reduce procurement complexity for organizations already committed to AWS spending.
How It's Different
- No infrastructure provisioning required: Unlike Oracle Exadata Database Service on Dedicated Infrastructure (also available on Oracle Database@AWS), ADB-S requires zero infrastructure setup — no Exadata VM clusters, no capacity planning, making it fundamentally simpler to start.
- Serverless vs. dedicated: Traditional Oracle Database@AWS offerings require provisioning dedicated Exadata infrastructure; ADB-S abstracts this entirely, offering a true serverless experience with per-use billing rather than reserved capacity charges.
- Autonomous self-management: ADB-S automatically handles patching, performance tuning, and index optimization via Oracle's Autonomous Database engine — capabilities not present in standard Oracle database deployments on AWS or even on Amazon RDS for Oracle.
- AI-first workload taxonomy: The four workload types (AI Transaction Processing, AI Lakehouse, AI JSON Database, Oracle APEX) are purpose-built and pre-tuned for their respective patterns, unlike generic database instances that require manual configuration.
- OCI-inside-AWS architecture: Unlike RDS for Oracle or self-managed Oracle on EC2, ADB-S runs on Oracle's own Exadata hardware physically located inside AWS data centers, providing full Oracle feature parity and Exadata performance with AWS-native connectivity.
- Cross-vendor HA with AWS storage: Autonomous Data Guard for HA combined with Amazon S3 for backups creates a hybrid resilience model that leverages the strengths of both Oracle and AWS infrastructure simultaneously.
When to Prefer It
- Migrating Oracle workloads with minimal refactoring: When an organization needs to lift Oracle Exadata or on-premises Oracle databases to AWS without rewriting applications or changing database features, ADB-S provides the highest compatibility path.
- Variable or unpredictable workloads: When database compute and storage demands fluctuate significantly (e.g., end-of-month batch processing, seasonal spikes), the serverless auto-scaling model avoids over-provisioning costs associated with dedicated infrastructure.
- Teams without Oracle DBA expertise: When the organization lacks deep Oracle administration skills, ADB-S's autonomous patching, tuning, and scaling removes the need for manual DBA intervention on routine operational tasks.
- Oracle APEX application development: When building low-code Oracle APEX applications on AWS, ADB-S provides the native, optimized runtime environment without requiring a full dedicated database cluster.
- AI and vector search workloads: When building generative AI applications that require Oracle AI Vector Search alongside transactional or analytical data, ADB-S provides the integrated Oracle 23ai feature set with low-latency AWS application connectivity.
- Consolidated AWS + Oracle spend: When an enterprise has AWS committed spend (EDP) and wants Oracle database costs to count toward that commitment while also earning Oracle Support Rewards, ADB-S through AWS Marketplace is the optimal procurement path.
- Rapid prototyping or dev/test: When developers need a fully functional Oracle Autonomous Database environment quickly without infrastructure lead times, ADB-S enables immediate provisioning from the AWS Console.
Availability
- GA status: Generally Available as of June 17, 2026.
- Supported regions: US East (N. Virginia) and US West (Oregon) only at launch; no additional regions announced.
- Licensing models: Supports both Bring Your Own License (BYOL) and License Included options, available via public offer or negotiated private offer on AWS Marketplace.
- Procurement: Available through AWS Marketplace; spend is eligible for AWS committed spend programs (e.g., EDP) and Oracle Support Rewards.
- Workload types: Four supported — AI Transaction Processing, AI Lakehouse, AI JSON Database, and Oracle APEX.
- HA and DR: Autonomous Data Guard included; automated backups to Amazon S3; cross-Region disaster recovery supported.
- AWS service integrations: AWS KMS (encryption), Amazon CloudWatch (monitoring), Amazon EventBridge (event management) are supported at launch.
- Limitation: Dedicated Exadata infrastructure provisioning is not required but also not available through the ADB-S serverless path; customers needing dedicated clusters should use the separate Exadata Database Service on Dedicated Infrastructure offering.
