Virtualisation Strategy: Rethinking Enterprise Infrastructure
Consultant
Nick Whitfield
For years, enterprise virtualisation strategy was relatively straightforward. Virtual machines became the default way to consolidate infrastructure, improve utilisation and gain flexibility without fundamentally changing the applications sitting above them.
That certainty is beginning to disappear.
Changing licensing economics, ageing infrastructure, cloud adoption, the rapid development of container platforms and increasingly demanding AI workloads are forcing organisations to ask a much bigger question:
Recent developments between Microsoft and Red Hat demonstrate just how quickly that landscape is changing.
The two companies have expanded their engineering collaboration to bring Red Hat OpenShift Virtualisation closer to the underlying compute capabilities of Microsoft Azure. The work includes technologies such as KubeVirt, LibVirt, QEMU and the Linux kernel, enabling organisations to run traditional VM-based workloads alongside containers and emerging AI workloads on a more unified cloud-native platform.
Significantly, Red Hat explicitly points to rising hypervisor licensing costs, hardware scarcity and increasing AI compute requirements as factors causing enterprises to reconsider where and how their virtual machines run.
This is another indication that enterprise virtualisation is entering a period of strategic reassessment.
For many organisations, the immediate reaction may be to compare platforms.
Should we remain with our existing virtualisation environment? Move more workloads to Azure? Adopt OpenShift Virtualisation? Move towards Kubernetes? Accelerate containerisation?
But those questions quickly expose a much more complicated reality.
Large enterprises rarely operate clean, uniform infrastructure estates. They have business-critical applications built across different generations of technology, supported by different vendors and internal teams, often spread across multiple geographies.
Some workloads may be ready for cloud migration.
Some might benefit from containerisation.
Others are perfectly suitable as virtual machines and simply need a more economical or flexible platform.
And some legacy applications may still be critical to the business despite being difficult, expensive or risky to migrate.
The danger is therefore treating the virtualisation reset simply as a technology selection exercise.
The temptation during any major technology shift is to search for a replacement.
But replacing one platform with another does not necessarily constitute modernisation.
The better starting point is the workload.
Organisations need to understand their existing estate, dependencies, support requirements, performance characteristics, security obligations and business criticality before selecting the destination architecture.
That assessment might lead to several outcomes.
A legacy VM could move largely unchanged into a new environment.
Another application might move onto Azure infrastructure.
Another could run through OpenShift Virtualization, creating a pathway towards a more cloud-native operating model.
Another might justify full containerisation.
And certain workloads may simply need to remain where they are until the economics and operational risk make migration worthwhile.
The expanded Microsoft-Red Hat collaboration is interesting precisely because it acknowledges this reality. Red Hat describes a model in which organisations can run VM-based applications alongside containers and AI workloads, providing greater freedom to modernise applications at their own pace rather than requiring everything to be transformed simultaneously.
For enterprise technology leaders, that flexibility is potentially extremely valuable.
It also makes the decision-making process more complex.
Once an organisation has selected its direction, another problem appears.
Who is going to deliver it?
A major virtualisation modernisation programme can rapidly require expertise spanning:
Few organisations maintain deep expertise across every one of those areas internally.
More importantly, those skills are not necessarily required indefinitely. An organisation may need significant specialist capability during assessment and migration, followed by a very different operating model once the new environment stabilises.
That creates an execution problem, not just a technology problem.
And this is where Penta believes organisations need to think differently about transformation.
At Penta Consulting, we see successful technology transformation as something that sits between strategy and execution.
Selecting the destination architecture matters. But achieving the business outcome depends on whether organisations can assemble the specialist expertise, delivery capacity, operational support and governance required to get there.
Penta already supports complex transformation across cloud, infrastructure, AI, legacy technology and large-scale international programmes, with a model deliberately designed around overcoming specialist capability shortages and operational complexity. Penta can support cloud migration and optimisation, infrastructure modernisation and supporting legacy environments as organisations transition towards newer platforms
In the context of virtualisation, we see that journey broadly across five stages.
1. Assess
Understand the existing virtualisation estate.
That means going beyond server counts and licensing. Organisations need visibility into workloads, dependencies, utilisation, criticality, security requirements, technical debt and future business requirements.
The objective isn't simply to identify what can move.
It is to determine what should move and why.
2. Decide
Build the business and technical case around the available options.
For one workload, remaining virtualised may make sense. Another might move to Azure. Another may be appropriate for OpenShift Virtualization. Others could justify deeper application modernisation or containerisation.
The right strategy may therefore involve several destinations rather than one.
3. Stabilise
Not everything can move immediately.
Business-critical legacy environments still need to remain resilient, secure and supported while transformation takes place.
That principle already sits at the heart of Penta's legacy support: protect today's environment while creating a controlled route towards modernisation, rather than forcing transformation onto an unrealistic timetable.
4. Migrate and modernise
This is where strategy becomes an execution challenge.
Migration waves need to be prioritised. Specialist engineering capability needs to be available when required. Dependencies have to be managed, disruptions minimised and applications tested before workloads move into production.
And where modernisation goes beyond infrastructure migration, organisations may need additional cloud, Kubernetes, platform engineering, automation and application expertise.
5. Optimise and operate
Migration should not be considered the finish line.
The new environment needs to deliver the economics, resilience, flexibility and performance that justified the transformation in the first place.
That requires continued optimisation, monitoring, platform expertise and clear operational accountability.
Penta does not believe every organisation needs the same virtualisation strategy.
What they do need is the ability to make an informed decision and execute it successfully.
Our role can therefore flex around the transformation itself.
Through Professional Services, Penta can introduce specialist technical capability for areas such as consulting, implementation and systems integration.
Through Managed Resource, organisations can access and deploy specialist capability across international markets while Penta manages the employment, compliance and administration complexity.
And through Managed Solutions, Penta can take greater responsibility for defined areas of delivery and ongoing technology operations.
Those three service models are already central to how Penta supports organisations facing specialist skills shortages, technology complexity and international delivery requirements.
This matters because the challenge isn't simply finding an engineer who knows a particular technology.
It is assembling the right capability at the right stage of the project.
Perhaps the biggest mistake enterprises could make is viewing the current shift purely through the lens of virtualisation licensing.
Cost may trigger the conversation.
But the opportunity is considerably larger.
If an organisation is reviewing its virtualisation platform anyway, it has an opportunity to reconsider how infrastructure supports its broader technology strategy.
Which applications genuinely need to remain as VMs?
Which could become cloud-native?
Which legacy workloads are consuming resources that could be better deployed elsewhere?
Where could automation reduce operational overhead?
And what infrastructure will be required to support the organisation's data and AI ambitions over the next five years?
These aren't questions about necessarily about replacing one hypervisor with another. But about building the right technology foundation for the next stage of the business.
The virtualisation landscape is changing, but enterprises should resist the temptation to rush towards whichever platform appears to be emerging.
Start instead with the estate.
Understand the workloads.
Model the options.
Identify the risks and dependencies.
Decide what should remain, migrate or modernise.
Then establish whether you have the people and delivery capability required to make that strategy real.
Because the biggest risk in the coming virtualisation reset isn't necessarily choosing the wrong technology.
It's making the right decision and discovering you don't have the capability to execute it.
That's the gap Penta Consulting is built to help close.