Bring compute to data: The next phase of multi-cloud strategy

Insights

• Multi-cloud success depends on how well applications, data, and cloud services work together.
• Application modernization and database transformation are separate architectural decisions.
• Re-platforming Oracle environments can add significant cost, complexity, and risk.
• Oracle Database@Azure, Oracle Database@AWS, and Oracle Database@Google Cloud help connect hyperscaler capabilities with existing Oracle environments.
• Organizations that start with data and application dependencies are better positioned to adopt AI and cloud innovation at scale.

For years, multi-cloud strategies focused primarily on application placement. Today, architecture teams are confronting a broader challenge: how to combine the strengths of multiple cloud ecosystems without fragmenting the applications, information, and processes that support enterprise operations.

Multicloud has become a mainstream operating model. Flexera's 2026 State of the Cloud Report found that hybrid cloud adoption has reached 73% of organizations, up 3 percentage points year over year, while 14% of organizations now operate exclusively in a multi-cloud environment without a private cloud. Organizations are increasingly distributing workloads across hyperscalers to access specialized capabilities in cloud-native services.

As environments become more distributed, complexity increases. Flexera's research points to this as a defining challenge of the current phase of cloud adoption, reflecting the difficulty of coordinating applications, data, governance, and operations across multiple platforms.

For many enterprises, this challenge is amplified by the fact that critical processes continue to rely on applications and databases that have evolved over decades. As cloud ecosystems become more diverse, successfully modernizing complex environments while maintaining architectural continuity is becoming a central consideration in multicloud strategy.

Balance innovation and continuity in a multicloud world

Enterprise technology environments are becoming increasingly diverse. Different hyperscalers offer distinct advantages. Azure is often chosen for enterprise integration and productivity ecosystems, AWS provides a broad range of cloud services, and Google Cloud is frequently selected for analytics and data-intensive workloads.

At the same time, the applications that support finance, supply chain, manufacturing, engineering, and customer operations continue to rely on information accumulated over years of business activity. Enterprise Resource Planning (ERP) platforms, financial applications, manufacturing systems, and engineering environments hold the transactions, records, and process knowledge that support day-to-day enterprise activity.

For many organizations, Oracle databases form a foundational layer within these environments. They manage customer orders, inventory movements, financial transactions, manufacturing operations, and supply chain activities. Over time, they have become deeply embedded in broader application landscapes through customizations, integrations, and operational workflows. As a result, Oracle remains central to many of the platforms that organizations use to execute and manage critical business processes.

Technology leaders therefore face a fundamental architecture challenge. They want to take advantage of innovations emerging across cloud ecosystems, including AI, analytics, application development, and industry-specific services. They also need to protect existing investments and maintain the reliability of established platforms. Increasingly, modernization success depends on how effectively organizations connect new cloud capabilities with the systems and operational foundations already embedded across the enterprise.

Misconceptions that complicate multicloud modernization

As organizations expand their multicloud strategies, they often discover that selecting a cloud provider is only part of the challenge. Equally important is determining how applications, data, and cloud services interact across multiple environments. Several common assumptions can make that task more difficult than expected.

Misconception 1: Multicloud modernization requires database re-platforming

A common assumption is that adopting Azure, AWS, or Google Cloud means Oracle databases must also be replaced, migrated, or converted to another platform. In reality, Oracle databases are often deeply woven into applications, workflows, and operating processes that have developed over many years.

Consider an ERP system that supports finance, procurement, inventory management, and supply chain operations. The database does far more than store information. It supports application logic, business rules, integrations, reporting processes, and operational workflows. What initially appears to be a database migration can quickly expand into application changes, integration redesign, code modifications, testing, process validation, and user retraining activities. Once these changes are made across multiple systems, reversing course can require another round of application updates, testing, and operational adjustments, increasing both cost and complexity. As a result, the effort involved can be significantly greater than anticipated.

Full re-platforming isn't impossible — organizations with enough engineering investment have done it. Amazon's own internal migration off Oracle is instructive precisely because of its scale: the project ultimately touched over 100 consumer-facing services, and involved moving nearly 7,500 Oracle databases holding 75 petabytes of data. Amazon's own account of the effort is candid that it was neither quick nor easy, even with a dedicated team of specialized engineers driving it. That's a useful data point for weighing the alternative: re-platforming is achievable, but the cost and timeline scale with how deeply the database is woven into surrounding applications — which is exactly the condition most enterprise Oracle environments are already in.

Misconception 2: Applications and databases can be separated without consequence

Another assumption is that applications can be moved to one cloud platform while databases remain elsewhere with little impact. While this might be acceptable for some workloads, challenges can emerge when applications depend on frequent interactions with underlying information.

For example, an order management application might be accessing inventory information, pricing data, customer records, and fulfillment status throughout the day. When applications and databases operate in different environments, every request must cross network connections before a response is returned. A single delay will be insignificant, but thousands of transactions executed across operational processes can cumulatively affect responsiveness and user experience.

As organizations introduce AI, analytics, and real-time decision support into these environments, timely access to information becomes increasingly important.

Misconception 3: Licensing and governance can be addressed later

Many multicloud initiatives focus first on infrastructure and application migration, leaving licensing, governance, and operational management for later stages. In practice, these considerations often influence both modernization costs and long-term sustainability.

Many enterprises have substantial investments in Oracle licensing programs, including Bring Your Own License (BYOL) and Unlimited License Agreements (ULAs). When licensing implications are overlooked early in a transformation program, organizations can end up paying for additional cloud database services while underutilizing existing investments, weakening the overall business case.

Governance presents a separate challenge. As applications, databases, and cloud services spread across multiple environments, teams must consistently manage security policies, access controls, monitoring tools, and compliance requirements. Delayed governance planning often increases operational complexity and makes it harder to maintain visibility into risk, cost, and performance — a challenge that reflects a broader pattern Flexera's research points to: management overhead, more than infrastructure cost, is increasingly what enterprises flag as multicloud's central complexity.

These misconceptions create a broader architectural challenge. Organizations want to take advantage of capabilities available across hyperscaler ecosystems while maintaining the reliability, performance, and continuity of existing environments. These challenges are prompting organizations to rethink a long-held assumption: that application modernization and database transformation must occur at the same time.

Bring compute closer to enterprise data

New multicloud architectures are emerging in support of this approach, allowing cloud adoption initiatives to move forward while preserving the platforms and information assets that already support key enterprise activities.

These architectures allow existing databases and modern cloud services to work together more closely, reducing the need for disruptive database transformation before modernization initiatives can begin.

Oracle Database@Azure, Oracle Database@AWS, and Oracle Database@Google Cloud are examples of this approach. They enable organizations to retain Oracle databases while accessing cloud services from Azure, AWS, or Google Cloud. Applications, cloud services, and databases can operate in close proximity, helping maintain performance while reducing the complexity associated with distributed architectures.

In practical terms, an organization might modernize a customer-facing application on Azure, adopt analytics services on Google Cloud, or use AI capabilities available through AWS while continuing to rely on the Oracle databases. This allows modernization initiatives to move forward without automatically triggering a large-scale database migration or re-platforming program.

The result is greater flexibility in how organizations adopt cloud services. Application teams can take advantage of the capabilities offered by different hyperscalers, while database teams continue to manage the environments that support core enterprise functions. Cloud innovation and existing enterprise systems can therefore evolve together, reducing transformation risk while creating a clearer path to modernization.

As part of its modernization program, a global technology company headquartered in Germany sought to move its Teamcenter and Test Data Management (TDM) applications to Microsoft Azure to improve scalability and support a distributed workforce. However, the underlying Oracle database was deeply integrated with application logic and operational workflows. Re-platforming would have introduced significant migration effort, testing complexity, and operational risk.

Infosys worked with the company to design and implement an Oracle Database@Azure architecture that enabled access to Azure infrastructure and services while preserving database compatibility, performance, and operational continuity.

By keeping applications and databases closely connected, the company advanced its modernization goals without disrupting the underlying environment. The results were measurable: application response times improved by 25%, while database cloning time was reduced by up to 90%, enabling faster provisioning of test environments.

Liantis, a Belgian provider of payroll and HR services, offers a different example of this approach. The company supports hundreds of thousands of self-employed workers and employers and was looking to strengthen its disaster recovery capabilities.

Liantis focused on improving the resilience of its technology environment through a disaster recovery modernization initiative. It moved its disaster recovery environment from Oracle running on standard Azure infrastructure to Oracle Database@Azure. According to the company's CTO, John Helsmoortel, the move simplified operations while maintaining fast connectivity between Oracle databases and Azure-based applications. This allowed technology teams to spend less time managing infrastructure and more time developing new products and services.

Oracle reports that the new environment delivered faster response times, stronger security controls, lower latency, and meaningful cost savings. The example highlights how organizations can use multicloud architectures to improve performance and operational resilience while continuing to build on existing Oracle investments. More broadly, organizations that evaluate application modernization and database transformation as separate architectural decisions can often achieve greater flexibility, lower risk, and a more sustainable path to innovation.

Start multi-cloud strategy with data, not platforms

Many multi-cloud strategies begin by asking which cloud provider is best suited for a particular workload. A more useful starting point is understanding where the information that supports core business processes resides and how it is connected to the broader application landscape.

Organizations should identify which systems hold the most important enterprise information, how closely that information is tied to applications and processes, and which workloads genuinely require transformation. Application modernization and database transformation often operate on different timelines and serve different objectives. Evaluating them independently can reduce risk, simplify execution, and accelerate time to value.

Technology leaders should also assess multi-cloud decisions through a broader lens that includes performance, licensing, governance, and operational complexity. Factors such as Oracle licensing investments, application dependencies, latency requirements, and security policies can have a significant impact on modernization outcomes. Addressing these considerations early helps avoid costly redesign efforts later.

Multicloud decisions should be guided by application dependencies, information architecture, governance requirements, and operational priorities, not simply by cloud platform selection. Those that align cloud adoption with the realities of their technology landscape will be better positioned to modernize at scale while maintaining operational continuity and long-term flexibility.

Conclusion

Multi-cloud strategy is entering a new phase. As cloud ecosystems continue to diversify, architectural complexity is increasingly determined by how information, applications, and services interact across environments.

Approaches such as Oracle Database@Azure, Oracle Database@AWS, and Oracle Database@Google Cloud enable organizations to keep Oracle databases closely connected to applications, AI services, analytics platforms, and other cloud capabilities. This helps reduce the complexity, latency, and operational challenges that can arise when applications and data are spread across multiple environments.

As organizations continue to expand AI, analytics, and cloud-native workloads across multiple cloud ecosystems, the number of interactions between applications and enterprise data will only increase.

Organizations that account for application dependencies, governance requirements, and information architecture early in their modernization journey will be better positioned to expand AI initiatives, adopt new cloud capabilities, and adapt to changing business priorities with greater confidence.