Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Cloud architecture in 2025 was less about choosing a provider and more about deciding where each workload belongs, how it will be governed and what it will cost to operate. AI capacity, regulation, resilience and cost accountability are reshaping those decisions—but the evidence points to more active redesign, not one winning architecture or a mass exit from public cloud.
What changed in cloud architecture in 2025?
“The cloud architecture renaissance” is a useful description of a shift in priorities, not the name of a formal movement or a settled industry consensus. Cloud decisions increasingly involve workload placement, AI capacity, data location, operational control, cost management and resilience alongside the familiar questions of provider and migration.
The evidence comes from surveys of different populations, in different regions, with different question wording. Their percentages should not be combined into a single adoption rate. Together, they show organizations revisiting architecture—not agreeing on one ideal design.
For example, Gartner forecast in May 2025 that 25% of organizations would have experienced significant dissatisfaction with cloud adoption by 2028, attributing the risk to expectations, implementation and/or uncontrolled costs. That is a forecast, not a measured dissatisfaction rate for 2025. Gartner’s cloud-trends announcement frames the broader shift as cloud becoming a business necessity and disruptor, rather than merely a technology enabler.
#1 Best Overall
How is AI changing cloud infrastructure?
AI makes capacity and workload placement more urgent. Gartner forecast in May 2025 that AI workloads would use 50% of cloud compute resources by 2029, up from less than 10% at the time of its announcement. This is a projection, not a measured 2025 outcome. Gartner also cautioned that organizations may need to bring AI to the data, rather than assume all data should move to a cloud service.
That placement question depends on the workload: data location and access rules, latency, accelerator availability, security, integration and economics can point in different directions. A public-cloud service may suit one use case; another may be better served where its data or specialized infrastructure already resides. Neither the forecast nor the surveys establish that every AI workload belongs in public cloud.
Separate AI interest from production readiness
In PwC’s 2025 EMEA Cloud Business Survey, 86% of surveyed organizations said agentic AI capabilities were decisive in provider selection, while 29% said they were scaling agentic AI. PwC surveyed 1,415 business and technology leaders across 26 EMEA territories between July and September 2025. Provider-selection criteria, exploration and scaled deployment are different measures; the gap does not mean that 86% had deployed agentic AI.
Recommended Free Tools
Rackspace Technology reported that 84% of respondents had taken steps to integrate AI and cloud strategies. Its survey was conducted by Coleman Parkes Research in October and November 2024 among 1,420 IT decision-makers and commissioned by Rackspace. “Taken steps” indicates activity, not production maturity or successful outcomes. Rackspace’s announcement also discusses hybrid cloud and AI integration as innovation drivers.
Rank #2
Is multicloud worth it, and when does hybrid cloud fit?
Hybrid cloud generally means using a combination of cloud environments and on-premises infrastructure; multicloud means using services from more than one cloud provider. Organizations may use either pattern to place workloads where they fit operationally, technically or legally. But the added environments also create integration, interoperability, security, cost-management and skills demands. The terms do not by themselves guarantee portability, resilience or savings.
Survey findings show interest in these approaches, but they measure different things. Foundry’s 2025 survey of 670 global IT decision-makers, fielded in May and June 2025, found that 39% were actively deploying or had fully deployed hybrid cloud, while 33% were researching or evaluating hybrid/multicloud solutions. Gartner forecast in May 2025 that more than 50% of organizations would not get expected results from multicloud implementations by 2029, citing the difficulty of connecting providers and lack of interoperability. A multicloud design needs a specific use case and a credible integration plan, not just a second provider.
Compare workload placement by the constraints that matter
| Approach | What it can help address | Questions and trade-offs to test |
|---|---|---|
| Public cloud | Use of provider-operated services and capacity without operating every underlying system yourself. | Can the service meet the workload’s data-location, access, latency, security and cost requirements? What happens if its dependencies or connectivity fail? |
| Private cloud or on-premises | Keeping selected workloads in an environment operated for one organization or within its own facilities. | Can the organization provide the required capacity, maintenance, resilience and specialist skills? Compare total operating cost, not just cloud bills. |
| Hybrid cloud | Keeping some workloads in one environment and others in another, including a mix of public cloud and on-premises or private infrastructure. | How will identity, networking, data flows, security controls, monitoring and recovery work across the boundary? |
| Multicloud | Using services from multiple cloud providers where distinct workload needs justify doing so. | What must interoperate or move between providers? Account for duplicated tooling, operational complexity, provider-specific dependencies and staff capability. |
| Sovereign cloud arrangement | Designing around defined requirements for data location, operational control, access and applicable jurisdiction. | What does “sovereign” mean for this workload? Verify the specific service, operator, access model and legal obligations rather than relying on a broad label. |
Gartner recommends identifying specific use cases and planning for distributed applications and data that could benefit from cross-cloud deployment. That is a more disciplined starting point than treating multicloud as a goal in itself.
Why are sovereignty and regulation shaping architecture?
Data location and control have become architectural requirements for many organizations, particularly in EMEA. PwC found that 82% of surveyed organizations in its 2025 EMEA Cloud Business Survey were refining their cloud approach in response to geopolitical and/or regulatory change.
Rank #3
Make “sovereignty” concrete before selecting a service. Specify where data is stored and processed, who can access it, which entity operates the service, which laws may apply and what control the customer retains. Then assess those requirements against the actual service and workload. A sovereignty label alone does not establish that a system meets a particular organization’s obligations.
These survey findings describe reported priorities; they are not legal advice or a determination of compliance with the EU AI Act, DORA, NIS2 or any national rule. The relevant obligations depend on the organization, workload and jurisdiction.
Why are some organizations moving workloads back?
Cloud strategy now includes selective repatriation—moving or considering moving some workloads to on-premises or private environments—as well as expansion. This does not establish that organizations are broadly abandoning public cloud.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Foundry reported that 75% of its 2025 survey respondents had moved or planned to move workloads back on-premises. That combined measure includes plans, not only completed moves; reported reasons included security, cost, reliability and compliance. Separately, more than two-thirds (69%) of respondents in Rackspace’s commissioned 2024 survey had considered repatriating at least some workloads from public cloud. “Considered” is not the same as having moved them, so the figures are not directly comparable.
The practical question is whether a workload’s operating costs, performance, control requirements and dependencies justify changing its location. Cloud is not intrinsically cheaper, and repatriation does not automatically save money. Include migration effort and the cost of operating the destination environment in the comparison.
How should teams manage cloud cost and value?
Cloud spending can grow without clear accountability for what it delivers. PwC reported that one in ten organizations had fully integrated advanced FinOps practices. FinOps is an operating discipline that connects financial accountability with workload performance; it is not simply a cost-cutting exercise.
For each material workload, teams should be able to relate consumption to an owner, a business purpose and a performance or service requirement. Review cost alongside latency, reliability, utilization and delivery needs. A lower bill that degrades a critical service is not necessarily a better architecture, and a flexible cloud bill without ownership can obscure poor value.
The dissatisfaction forecast from Gartner is a warning about expectations, implementation and/or uncontrolled costs—not proof that cloud adoption itself fails. Set measurable outcomes before migration or redesign, and revisit the assumptions as workload demand changes.
Best Value
What do cloud-native adoption and skills mean for architecture?
Architecture depends on the ability to operate it. CNCF’s Cloud Native 2024 report, published April 1, 2025, drew on 750 cloud-native community members surveyed in fall 2024. One-quarter of respondents said nearly all their development and deployment used cloud-native techniques. This describes a community survey, not a representative census of all enterprises.
Cloud-native methods and tools can form part of a design, but the survey does not show that every organization needs Kubernetes or serverless. Skills, integration and operational ownership should be considered alongside technology choices. Rackspace’s survey findings likewise concern reported activity and demand, not proof that any particular platform is the right fit.
How do you choose an architecture for a workload?
Start with the workload and its constraints, not a blanket preference for public, private, hybrid, multicloud or sovereign cloud. Record the requirements and identify which environment can meet them with acceptable operational complexity.
Free tools Windows power users keep installed
One-click scans. No signup required.
- Define the outcome. State what the workload must do and how success will be measured, including performance, reliability and business value.
- Map its data and dependencies. Identify data location and access requirements, latency sensitivity, integrations, connectivity needs and dependencies on provider-specific services.
- Set control and resilience requirements. Specify security responsibilities, applicable jurisdictional constraints, recovery objectives and what should happen if a provider or connection is unavailable.
- Compare lifecycle cost and capability. Include consumption, integration, migration, ongoing operations and the skills needed to support each candidate design. Tie cost ownership to service outcomes.
- Choose the least complex design that meets the requirements. Add another provider or environment only when a defined workload need justifies the additional integration and operating burden.
- Review against evidence from operation. Reassess placement when demand, regulation, service performance, costs or business needs change.
For EMEA organizations, PwC’s 2025 survey offers evidence of active adjustment: 94% of surveyed organizations planned to adjust and expand cloud architecture and coverage. That intention is not a guarantee of execution or of better outcomes. It reinforces the need to treat architecture as an ongoing set of workload-level decisions rather than a one-time migration destination. PwC’s 2025 EMEA Cloud Business Survey was published November 7, 2025.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
