Oracle and AWS Multi-Cloud Alliance Accelerates: AI Database Expansion and Interconnect GA Reshape Cross-Cloud Architecture

Introduction

This week, Oracle Cloud Infrastructure (OCI) announced that Oracle Interconnect for AWS is now generally available (GA). The service first debuted in limited availability in May 2026, and now provides production-grade managed connectivity in the OCI Ashburn and AWS US East (Northern Virginia) regions. At the same time, Oracle AI Database@AWS continues its global expansion, adding regions such as Stockholm, San Jose, São Paulo, and Madrid, bringing the total number of available regions to 22. These developments mark a shift of multicloud from an edge option in enterprise IT to a core architecture, while the depth of collaboration between cloud providers has reached unprecedented levels.

Background

Over the past year, the multicloud collaboration between Oracle and AWS has deepened progressively. In June 2025, Oracle AI Database@AWS became generally available, allowing enterprises to run Oracle database services in the AWS environment while maintaining unified AWS billing. On June 17, 2026, at AWS Summit New York, the two companies announced several milestones: Oracle Autonomous AI Database Serverless became generally available, Oracle Exadata Database Service on Exascale was announced as coming soon, regional coverage expanded to 20 AWS Regions, and a long-term strategic collaboration agreement (SCA) was signed. One week later, Oracle AI Database@AWS added the Stockholm, Sweden and San Jose, United States regions, bringing the total number of regions to 22. In addition, Oracle Interconnect for AWS was released in limited availability on May 15, and reached GA on July 29, providing secure, private, low-latency connectivity for cross-cloud traffic.

On the Google Cloud side, OCI GoldenGate also became generally available on May 20, providing real-time data replication capabilities for Oracle AI Database@Google Cloud, supporting multicloud data synchronization and reducing migration risk.

Technical Analysis

Oracle Interconnect for AWS的核心价值在于“托管互联”。传统跨云连接通常需要企业自建IPsec VPN或通过第三方专线提供商，成本高且配置复杂。Oracle Interconnect for AWS由Oracle和AWS联合构建和运营，使用户可以在OCI和AWS控制台中直接创建和管理连接。连接默认启用IEEE 802.1AE MACsec标准加密，确保数据在物理层得到保护。服务采用OCI全球统一连接定价且免收数据传输费用，大幅提升成本可预测性。

Oracle AI Database@AWS则重新定义了数据库部署模式。该服务将OCI基础设施部署在AWS附近，使Oracle Real Application Clusters（RAC）、Exadata Database Service、Autonomous AI Lakehouse和AI Vector Search等能力以托管形式提供给AWS客户。客户无需重新设计应用，即可获得Oracle数据库的企业级性能，并直接在数据库中使用AI功能。例如，Autonomous AI Database Serverless允许企业基于实时数据运行AI训练和推理，无需将数据复制到独立的数据湖。

企业影响分析

对于长期在AWS中运行应用、又需要Oracle数据库的企业，这一变化直接解决了“两个云之间数据交换”的痛点。成本方面，Interconnect for AWS的统一低定价和免费数据传输，让跨云流量不再是财务黑箱。与租用专线相比，企业可节省大量网络基础设施的CAPEX和日常运维OPEX。部署方面，连接服务由两家云厂商协同运维，企业无需承担第三方合同和故障协调成本，应用交付周期得以缩短。

安全与合规层面，MACsec加密加上托管连接模型，减少了通过公共互联网传输的风险，有助于满足金融、医疗等行业的合规要求。区域扩张（如圣保罗、马德里、斯德哥尔摩）使企业可在本地就近部署，满足数据驻留和低延迟需求。对计划迁移的企业而言，Oracle AI Database@AWS使得从本地数据中心直接迁移到AWS成为可能，例如将Oracle RAC数据库原样运行在AWS环境中，同时继续使用AWS的管理工具和计费。

市场竞争分析Oracle's multi-cloud strategy is reshaping the competitive landscape of the cloud market. By establishing deep integrations with AWS and Google Cloud respectively, Oracle has effectively become the "data infrastructure layer" of the cloud era—regardless of which hyperscaler a company chooses, it can access Oracle database services. This puts pressure on Azure, even though Azure also offers Oracle database products; however, the joint operating models between AWS and Google Cloud with Oracle are more in-depth. AWS, by introducing Oracle databases, has strengthened its appeal to large enterprise customers, preventing them from switching to other clouds to support Oracle workloads. Google Cloud, after gaining OCI GoldenGate support, has also improved its competitiveness in data replication and hybrid architectures.

In addition, the GA of Oracle Interconnect for AWS means that multi-cloud network connectivity has become a standardized service rather than a custom IT project for enterprises. This diminishes part of the value that traditional network operators and SD-WAN providers hold in the cloud interconnect market, and reduces enterprises' reliance on third-party MPLS or dedicated lines. In the future, interconnections between cloud providers may become standard, and connectivity itself will no longer be a competitive differentiator.

Industry Trend Observations

From a broader perspective, these events reflect three long-term directions. First, multi-cloud is shifting from "avoiding vendor lock-in" to "orchestrating workloads across multiple clouds." Enterprises are no longer agonizing over which cloud to choose, but rather how to make multiple clouds work together to generate maximum value. Second, the convergence of AI and databases is accelerating. The emergence of products such as Autonomous AI Database Serverless means that AI capabilities are embedded directly into the transactional data layer, rather than being services independent of the data. This aligns with the trend of "AI-native databases," enabling enterprises to execute AI where data is generated, reducing data movement costs. Third, regional expansion is a key means of meeting regulatory and data sovereignty requirements. As data localization demands increase in regions such as the EU and South America, cloud providers are rapidly expanding available regions to help enterprises mitigate compliance risks.

CloudTechDaily InsightThe deep collaboration between Oracle and AWS is essentially a microcosm of the cloud industry shifting from "single-cloud competition" to "multi-cloud collaboration." The GA of Oracle Interconnect for AWS and the rapid expansion of AI Database@AWS have transferred the complexity of cross-cloud connectivity and data services from enterprises to cloud vendors themselves. This brings three implications for enterprise IT strategy: First, future architecture design no longer requires provisioning large amounts of network engineering resources for cross-cloud setups; managed connectivity allows teams to focus on applications and data value. Second, the multi-cloud deployment of database workloads is becoming a reality, and enterprises' cloud choices will be based more on upper-layer service capabilities than on underlying infrastructure. Third, the degree of integration between data and AI will become key to enterprise competitiveness. The ability to run AI on online transaction data, such as Oracle Autonomous AI Database Serverless, may deliver more practical business value than standalone AI platforms. Over the next five years, we expect multi-cloud data infrastructure to become the fastest-growing segment of enterprise IT spending, and the collaboration model among Oracle, AWS, and Google Cloud may become the mainstream paradigm for the entire industry.

---

*Source: Oracle official blog - Oracle Multicloud – What’s News*