Oracle Database@Azure Expands to West Europe and Brazil with Exadata on Exascale

Oracle’s multicloud play tightened its grip on Europe and Latin America this month as Oracle Database@Azure expanded into the Netherlands’ West Europe region and added capacity in Brazil, bringing new service options—most notably Exadata on Exascale and enhanced key-management integrations—to organizations that rely on Oracle workloads while building on Azure’s AI and analytics stack.

Background​

Oracle Database@Azure is a co‑location model: Oracle runs and manages Oracle Database services (Exadata Database Service, Autonomous AI Database, Base Database Service, GoldenGate, and related features) on Oracle infrastructure physically located inside Microsoft Azure datacenters. The idea is to preserve the full operational semantics and high‑availability features of Oracle databases while giving Azure applications, analytics, and AI tooling low‑latency access to transactional data. That architecture enables a practical multicloud strategy for enterprises that want to keep Oracle operational models intact while exploiting Azure’s AI, Microsoft Fabric/OneLake analytics, and governance services. Oracle and Microsoft have been steadily expanding region availability since the program’s launch, and the most recent announcements add both capacity and capabilities: Oracle Exadata Database Service on Exascale Infrastructure (a hyper‑elastic, pay‑per‑use Exadata option) is now available on Database@Azure, a Base Database Service SKU for virtual‑machine hosted Oracle Database is being introduced, and Azure Key Vault integration for customer‑managed TDE keys has been made broadly available for certain database SKUs. These items change the procurement, encryption, and scaling calculus for cloud database owners.

---

What exactly changed: the short list​

• New regions added: Oracle Database@Azure is now available in West Europe (Netherlands) and received expanded capacity in Brazil (Brazil Southeast complementing existing Brazil South), widening local options for EMEA and South America deployments.
• Exadata on Exascale is GA on Database@Azure: Oracle announced Exadata Database Service on Exascale Infrastructure for Oracle Database@Azure, enabling elastic, pay‑per‑use Exadata without dedicated infrastructure provisioning.
• Base Database Service: A simpler VM‑based managed database SKU (Oracle Base Database Service) will appear for standard/entry workloads with pay‑as‑you‑go pricing and reduced administration overhead.
• Azure Key Vault for TDE master keys: Integration with Azure Key Vault (Standard, Premium, Managed HSM) is now available so customers can store and manage Transparent Data Encryption (TDE) keys under Azure control.
• Stronger Fabric/OneLake mirroring and GoldenGate integrations: Open Mirroring (public preview) and GoldenGate enhancements simplify near‑real‑time replication into Microsoft Fabric and OneLake for analytics and AI workloads.
• Rapid regional expansion plan: Public announcements cite the offering live in more than 30 Azure regions with plans to bring the total to 31–33 regions over ongoing rollouts; region counts vary by vendor statement and documentation at the time of writing. Validate region‑SKU availability for specific workloads before planning production deployments.

---

Why the Netherlands and Brazil matter​

West Europe (Netherlands): a strategic European anchor​

The West Europe (Amsterdam) Azure region is one of Europe’s key connectivity hubs and a frequent selection for organizations needing EU data residency, low latency to Northern/Western Europe, and proximity to major financial and logistics customers. Oracle’s move to make Database@Azure available in West Europe closes a geographic gap for enterprises that want Oracle operational models in Azure datacenters inside the EU, removes unnecessary latency to Azure AI and Microsoft Fabric services for Amsterdam‑hosted apps, and supports European compliance and disaster‑recovery topologies. Oracle’s regional expansion to West Europe also positions the company to support enterprises responding to sovereign cloud requirements and EU regulatory scrutiny over cross‑border data flows.

Brazil: South America capacity and DR resilience​

Brazil has become a crucial growth region for cloud services, driven by digital transformation in finance, retail, telecoms, and public sector projects. Oracle Database@Azure’s presence in Brazil South (São Paulo) and Brazil Southeast (Rio) provides organizations in Latin America with local Exadata performance options and the ability to design resilient, low‑latency topologies with adjacent DR regions. For multinational companies and MSPs serving Latin America, local availability reduces network egress costs, improves application responsiveness, and simplifies compliance with Brazilian data‑locality expectations.

---

Technical deep dive: what the services deliver (and what to verify)​

Exadata Database Service on Exascale Infrastructure​

Exadata Exascale is Oracle’s elastic Exadata offering designed to bring Exadata performance to a pay‑as‑you‑go, multi‑tenant model. It uses pooled, RDMA‑capable storage and distributed Smart Scan algorithms to accelerate queries and deliver the familiar Exadata feature set without minimum dedicated-server reservations.
What Oracle claims:

• Up to 95% lower minimum infrastructure costs for entry-level use compared with dedicated Exadata nodes.
• Elastic scaling with no per‑IO charges and intelligent distributed storage algorithms that accelerate queries.

What to verify:

• Benchmark claims (for example, vendor microsecond I/O latency figures) originate from vendor tests under specific workloads and hardware topologies. Real‑world latency and throughput are influenced by network topology, private interconnect setup (FastConnect/ExpressRoute), database chatty‑ness, and application design. Treat microsecond storage claims as vendor benchmarks—run a proof‑of‑value (PoV) with representative workloads to validate expected latency and throughput.

Oracle Base Database Service​

Oracle Base Database Service brings the Oracle Database engine to a VM‑based, managed SKU inside Azure datacenters on Oracle infrastructure. It supports Oracle Database Enterprise Edition and Standard Edition 2 (19c and 23ai), provides lifecycle automation, and is positioned for pay‑as‑you‑go workloads that do not require Exadata‑class performance. Practical implications:

• Easier administration for smaller workloads or dev/test environments.
• Potentially different SLAs and performance characteristics compared with Exadata SKUs—validate service limits, availability zones, and backup/restore behaviours against production requirements.

Key‑management: Azure Key Vault + TDE​

Customer‑managed keys for Transparent Data Encryption (TDE) through Azure Key Vault are now generally available for some Database@Azure SKUs. This integration centralizes cryptographic control within Azure while Oracle handles the database layer, enabling unified key rotation, HSM support, and consolidated audit trails. What to test:

• Key rotation and recovery workflows across Oracle and Azure control planes.
• HSM usage patterns (Managed HSM vs. Premium KV) and associated SLAs.
• Backup encryption and cross‑region key policy behaviour during failover.

GoldenGate, Open Mirroring and Microsoft Fabric/OneLake​

Oracle has improved GoldenGate integrations and introduced Open Mirroring to replicate Oracle tables into Microsoft Fabric/OneLake in near real time, producing analytics‑ready Parquet/Delta tables for Power BI, Fabric analytics, and Azure AI. This addresses the long‑standing ETL problem for operational analytics and AI by enabling fresher, governed copies of Oracle data for lakehouse pipelines. Operational caveats:

• Mirroring and CDC pipelines must be validated for schema drift, latencies, and transactional consistency depending on analytic use cases.
• Licensing and GoldenGate transaction volumes may materially affect cost. Confirm licensing with procurement.

---

Commercial and contractual considerations​

Oracle Database@Azure introduces cross‑vendor complexity: Oracle manages the database layer while Azure controls the datacenter networking, billing, and many surrounding services. This model creates operational advantages and contractual risks.
Key procurement points:

• SKU and pricing clarity: Confirm exactly which Oracle SKU is available in your target Azure region (Exadata Dedicated, Exascale, Autonomous AI Database, Base Database Service) and which pricing model applies (BYOL, license‑included, pay‑as‑you‑go). Vendor announcements have shown evolving SKU availability by region; count and names have changed over announcements—verify the region matrix during procurement.
• Billing and marketplace mechanics: Understand whether consumption flows through Azure Marketplace or is invoiced by Oracle, and how Azure commitment credits or reserved capacity apply.
• SLA and incident response: Define escalation, SLOs, and cross‑vendor incident responsibilities in the contract. Confirm backup/restore and DR behaviours across Oracle and Azure control planes.

---

Security, compliance and governance — practical realities​

Oracle and Microsoft highlight integrations with Microsoft Purview, Entra ID, Defender for Cloud, and Sentinel to provide a full governance and security stack across the Oracle/Azure boundary. The Azure Key Vault integration for TDE is a critical step because it centralizes key custody within the tenant’s Azure footprint. What security teams should validate:

• End‑to‑end audit trails spanning Azure and Oracle control planes.
• Purview lineage and classification for mirrored OneLake datasets.
• Identity flows and least‑privilege access models for Oracle management APIs that interact with Azure services.
• Regulatory compliance mapping for regional deployments (e.g., GDPR obligations for EU deployments in West Europe).

---

Strengths: where this alliance clearly helps enterprise IT​

• Low‑latency multicloud: Co‑location reduces the round trip between Azure compute/AI and Oracle transactional systems, enabling near real‑time AI use cases and analytics that previously required re‑engineering or heavy ETL.
• Operational continuity for Oracle customers: Enterprises can preserve Oracle RAC, Data Guard, and advanced availability features while exposing data to Azure services. That reduces migration risk for mission‑critical systems.
• Improved governance options: Azure Key Vault support and Purview integrations make centralized key and data governance easier across co‑located Oracle databases and Azure analytics stacks.
• Flexibility and cost choices: Exadata Exascale and Base Database Service widen the spectrum from hyper‑performance Exadata to smaller, VM‑based deployments—potentially lowering entry costs for smaller customers.

---

Risks and limitations: the caveats IT leaders must not ignore​

• Cross‑vendor complexity: Operational responsibility is split. Patching, incident response, and DR procedures likely involve Oracle and Microsoft teams; clarify contact points and SLAs. Lack of clarity creates risk for production workloads.
• Hidden costs: GoldenGate licensing, data‑transfer charges, and marketplace mechanics can repeatedly surface costs—confirm cost forecasts under representative loads.
• Feature parity by region: Not every Oracle SKU or capability is available in every Azure region simultaneously. Region counts and available SKUs have changed across vendor announcements; treat region lists as dynamic and verify SKU availability in target regions before committing.
• Performance claims need empirical validation: Vendor latency/throughput benchmarks (for example, Exascale microsecond I/O claims) are helpful as an upper bound but are not a substitute for PoV testing under real application behaviour.
• Governance surface area expands: Mirroring Oracle data into OneLake enables analytics but also increases the governance burden: cataloging, lineage, and retention rules must be aligned across Oracle and Fabric pipelines.

---

A practical checklist for IT teams evaluating Oracle Database@Azure​

• Validate region‑level SKU availability for your target Azure regions and SKUs. Confirm whether Exadata Dedicated, Exascale, Autonomous AI Database, or Base Database Service is offered there.
• Run a proof‑of‑value (PoV) that includes representative workloads and measures latency, failover behaviour, backup/restore times, and GoldenGate/CDC pipeline freshness under load.
• Confirm key management: test Azure Key Vault HSM workflows, key rotation, and cross‑region key recovery during simulated DR events.
• Map the security and governance model end‑to‑end: ensure Purview cataloging, Sentinel alerting, and Entra ID mapping are implemented and tested for both operational and mirrored analytic datasets.
• Determine procurement and billing mechanics: marketplace invoicing, BYOL eligibility, and any Azure commitment interactions that affect TCO.
• Define support and escalation: sign clear SLAs with joint Oracle‑Microsoft support playbooks for failover, performance incidents, and security breaches.
• Validate GoldenGate and mirroring for schema drift, latency, and downstream analytics compatibility (Parquet/Delta formats for OneLake/Fabric).

---

SEO‑friendly summary of technical takeaways​

• Oracle Database@Azure expansion to West Europe (Netherlands) and enhanced Brazil coverage provides local, low‑latency options for organizations that need Oracle database performance integrated with Azure AI, Microsoft Fabric, and OneLake.
• Exadata on Exascale makes Exadata performance elastic and pay‑per‑use; validate vendor latency claims with a PoV.
• Azure Key Vault integration enables centralized key control for TDE across Oracle Database@Azure SKUs.
• GoldenGate + Open Mirroring simplify near‑real‑time replication into Fabric/OneLake for analytics and AI.
• Region counts and SKU availability are dynamic—confirm what is available in the target Azure region before procurement.

---

Final analysis: who should care and the strategic call​

Large enterprises that operate significant Oracle estates but want to accelerate AI and analytics workloads on Azure benefit most from Oracle Database@Azure’s expanded European and South American footprint. The offering lowers the friction of combining Oracle’s transactional strengths with Azure’s AI and Microsoft Fabric analytics, enabling new classes of near‑real‑time applications for fraud detection, dynamic pricing, and governed AI model training.
However, the model is not a silver bullet. It introduces contractual and operational complexity that demands careful engineering, procurement discipline, and governance work. Teams that rush to production without realistic PoV testing, contractual clarity on SLAs and costs, and a robust governance pipeline for mirrored analytic datasets risk facing surprise costs and operational headaches.
For practical staging: start with a narrowly scoped pilot that exercises the full end‑to‑end path—transactional workload on Oracle, replication into Fabric/OneLake, and an Azure AI/Power BI consumption layer—validate SLAs and costs, and then iterate outward to broader production use.
Oracle and Microsoft’s announcements show deliberate progress: more regions, richer integrations (Key Vault, Purview, GoldenGate/Open Mirroring), and new price/performance options (Exascale, Base Database Service). That combination makes Oracle Database@Azure an increasingly credible multicloud option for enterprises where Oracle databases are core to business operations and Azure is the preferred platform for analytics and AI. But the success of any migration or new deployment will depend on disciplined validation, governance, and contractual clarity—non‑negotiable steps to turn vendor promises into production reliability.

---

Oracle Database@Azure’s Netherlands and Brazil expansions mark another step in the pragmatic multicloud era: high‑performance database capability brought close to cloud AI and analytics, with governance and security controls that reflect enterprise needs. The technical potential is clear; the operational work remains the decisive factor.

---

Source: ERP Today Oracle Database@Azure expands to the Netherlands, launches in Brazil