Marcus Muldoon
Head of Technical Services
August 17, 2026
The decisions that shaped an organisation’s IT estate five years ago may no longer make sense today. Virtualisation costs have shifted, AI projects are exposing gaps in data and infrastructure, and cyberattacks have made recovery just as important as prevention. Together, these pressures are prompting IT leaders to take a fresh look at how their technology environment is built, managed and protected.
From the conversations we are having with customers, three shifts stand out as having a particularly significant impact on IT strategy in 2026.
1. Organisations are reassessing virtualisation
Changes across the virtualisation market have prompted many organisations to examine platforms and licensing arrangements that may have otherwise remained largely unchanged for years.
We find that cost is often the initial trigger, but the conversation quickly becomes broader. IT teams are also considering how much flexibility their current environment provides, whether they are overly dependent on one platform and how effectively they can manage workloads across on-premises infrastructure, private cloud and public cloud.
This is encouraging a more considered approach to hybrid cloud. Rather than placing workloads in one environment by default, organisations are looking at where each application can deliver the best combination of performance, control, security and cost efficiency.
The objective should be to create an operating model that gives the organisation greater choice, flexibility and control – rather than replacing technology for the sake of it.
Solutions such as HPE Morpheus VM Essentials can provide a practical route into this transition, helping organisations run and manage virtual machines while reducing their reliance on a single virtualisation provider. Wider capabilities including HPE Morpheus Enterprise and HPE OpsRamp can then support more consistent provisioning, governance and visibility across hybrid environments.
The strongest strategies will begin with a clear assessment of the existing estate: which workloads are running, what they cost, how critical they are and where they should live in the future. This helps turn a licensing challenge into an opportunity to simplify operations and build a more adaptable foundation.
2. AI is moving from experimentation into operations
Most organisations now recognise the potential of AI. The greater challenge is moving from individual tools and isolated pilots to secure, repeatable use cases that deliver measurable value.
That transition places new demands on infrastructure.
AI applications need access to trusted, well-governed data. They may also require significant compute capacity, high-performing networks and the ability to process information across data centres, cloud platforms and edge locations.
Many AI projects stall because these foundations have not been addressed. Data can be fragmented across different platforms, infrastructure may not be designed for the workload and governance decisions are sometimes left until late in the project.
-
Organisations therefore need to think beyond the AI application itself. Before scaling a use case, they should understand:
- What business outcome is the project expected to deliver?
- Is the necessary data accessible, accurate and appropriately governed?
- Where should processing take place?
- Can the network and infrastructure support the required performance?
- How will sensitive information remain protected?
- What will it cost to operate the solution at scale?
There is unlikely to be one deployment model that suits every AI workload. Some applications may be appropriate for public cloud services, while those involving sensitive data, predictable high demand or sovereignty requirements may benefit from private or on-premises infrastructure.
-
HPE Private Cloud AI, developed with NVIDIA, is one example of how organisations can create a controlled environment for developing and deploying AI. HPE’s storage, compute and networking portfolio can also help address the wider requirements around data preparation, performance and connectivity.
The focus should remain firmly on the outcome. Successful AI adoption is not about building the largest possible platform. It is about creating the right foundation for valuable use cases and providing a clear route from pilot to production.
3. Cyber resilience is becoming a design principle
Preventing cyberattacks remains the first line of defence. Organisations need strong security controls that reduce vulnerabilities, protect users and data, and stop threats before they can cause disruption.
However, even the strongest defences cannot remove every risk. True cyber resilience combines prevention with the ability to contain an incident, maintain essential services and recover critical data and systems quickly.
This means organisations must consider both sides of the equation: how effectively they can prevent an attack and how confidently they could continue operating if one succeeds.
This is driving a shift from cybersecurity alone towards wider operational resilience.
-
Backup remains part of that picture, but having a copy of the data does not automatically guarantee recovery. Organisations need to know that recovery data is isolated, protected from tampering and available within the timescales the business requires. They must also understand the order in which systems will be restored and whether recovered data is clean. Technologies such as HPE Zerto can support continuous data protection and rapid recovery across virtualised and hybrid environments. HPE StoreOnce and HPE Alletra Storage MP can add further protection and availability, while security built into HPE ProLiant Compute helps establish trust at the infrastructure level.
-
Resilience also extends into the network. As users, devices, applications and data become more distributed, networking and security can no longer be treated as separate concerns. Secure access, consistent policies and greater visibility are becoming fundamental to both user experience and business continuity. Regular recovery testing is particularly important. A recovery plan that exists only as a document provides limited reassurance. Testing reveals dependencies, exposes gaps and gives teams the experience needed to respond confidently under pressure.
One connected infrastructure strategy
Virtualisation, AI and cyber resilience may appear to be separate areas of investment, but they increasingly depend on the same underlying decisions.
An open and manageable hybrid environment makes it easier to place workloads appropriately. A modern data foundation supports AI while improving governance and protection. Resilient infrastructure allows the organisation to innovate without introducing unacceptable risk.
This is why point solutions alone rarely provide the complete answer. Organisations need a joined-up view of their applications, data, infrastructure, networks and operational priorities.
At DTP, we help customers assess where they are today, identify the areas creating unnecessary cost or risk and build a practical roadmap for modernisation. As an HPE partner with deep expertise across hybrid cloud, infrastructure, networking and cyber resilience, we can bring the relevant capabilities together around the organisation’s specific requirements.
The technology landscape will continue to change. The priority for IT leaders is to build an environment capable of changing with it: one that provides greater choice, supports AI where it can deliver genuine value and keeps the organisation operating when disruption occurs.
