ChatGPT
Perplexity
Claude
Two years after Broadcom closed the VMware acquisition, the consequences are no longer a procurement conversation. They are an infrastructure strategy conversation, and they reach into decisions that VMware customers thought were settled five years ago.
The licensing shift has accelerated cloud migrations that were already on the roadmap, paused others that no longer make economic sense, and put hypervisor alternatives on the agenda for the first time in a decade. This post looks at what has actually changed in 2026, why CIOs are reorganizing infrastructure plans around it, and what the patterns look like across enterprise segments.
What changed in VMware licensing
A short recap, because the public information has shifted four or five times since 2023 and the original announcements are now mostly outdated:
- Late 2023: Broadcom completes the VMware acquisition and signals an aggressive subscription transition
- Early 2024: Perpetual licenses are discontinued. All customers move to subscription via VMware Cloud Foundation (VCF) or Vsphere Foundation (VSF), the two bundled SKUs
- Mid-2024: Channel program changes shrink the partner ecosystem. Smaller resellers lose access, mid-market customers find fewer paths to negotiated pricing
- 2024 to 2025: Public reports and analyst estimates put typical renewal increases at 3x to 5x, with a tail of 6x to 10x for customers on the worst-fit bundles
- 2025: VMware Cloud on AWS is wound down for new customers. AWS launches Amazon Elastic VMware Service (EVS) as the bring-your-own-license replacement
- 2026: Most enterprises are now past their first renewal under the new model. The strategy questions have stopped being hypothetical
The shift is not just a price increase. It is a structural change in how VMware is sold, what it is bundled with, and how organizations consume it.
Why the bundling matters
VMware customers used to buy components. Some bought vSphere only. Some added vSAN. Some added NSX. Each piece had its own price, and customers paid for what they used.
The new SKUs bundle these together. A customer who used vSphere alone now pays for vSAN and NSX whether or not they run them. A customer who used a fraction of the VCF stack now pays for the whole stack. The price increases that get headlined in analyst reports are often the visible part of a more uncomfortable reality: many customers are also paying for components they will never use.
This is the part of the change that turns a procurement problem into a strategy problem. When the unit economics of the software shift, the architecture decisions that depended on those economics shift with them.
The strategic shifts happening in 2026
Across the migration and consulting engagements happening this year, five patterns keep showing up.
1. Modernization timelines compressed
Enterprises that had cloud-native roadmaps stretched across 5 to 7 years are now running 18 to 36 month plans. The reason is straightforward: a Broadcom renewal at 4x the prior cost makes a 24-month modernization payback look attractive, even when the same project at the old cost did not pencil out.
The economic flip is significant. When VMware was a stable, predictable cost, modernization had to justify itself against a competitor that was cheap and known. When VMware becomes expensive and uncertain, modernization is justified just by capturing what would otherwise be paid to Broadcom.
For the wider replatform-vs-rehost decision, the 6Rs cloud migration framework is the standard model and has been adapted in many engagements to weight the timeline trade-offs against renewal pressure.
2. Hypervisor lock-in becomes a board question
For most of the 2010s and early 2020s, the hypervisor was not a strategic decision. VMware ran everywhere, and the cost of switching was higher than the cost of staying. Cloud-native organizations had Kubernetes and serverless conversations, but the existing virtualized estate stayed on vSphere.
In 2026, hypervisor choice is on the agenda again. The alternatives gaining serious evaluation:
- Nutanix AHV. The most enterprise-ready alternative. Strong vSphere migration tooling, hyperconverged architecture, growing in mid-market and enterprise accounts. Has its own commercial trajectory, so customers replacing Broadcom dependency with Nutanix dependency are not eliminating risk, only changing vendors.
- OpenShift Virtualization (KubeVirt). Red Hat’s Kubernetes-native approach to running VMs alongside containers. Compelling for organizations that already see Kubernetes as the platform of record.
- Proxmox VE. The open-source dark horse. Gaining traction in mid-market and education segments. Lower-cost commercial support tier. Lacks some of the enterprise features VMware customers take for granted, but the gap is closing.
- XCP-ng. A smaller community-driven alternative, mainly for organizations with existing Citrix Hypervisor footprints.
None of these are 1:1 replacements for vSphere. Each shifts the operating model, the skill set required, and the tooling around backup and DR. The fact that they are now in active enterprise evaluation, rather than dismissed as niche, is itself a signal of how much the conversation has changed.
3. Multi-cloud rebalances around new economics
Organizations that had been multi-cloud as a strategy for resilience or negotiation leverage are revisiting the calculus. The reasons:
- Egress between clouds is still expensive. When VMware was the common runtime everywhere, hybrid and multi-cloud were achievable without rebuilding networking. Without VMware as a connector, cross-cloud architectures require more deliberate design.
- Azure VMware Solution and Amazon EVS are not equivalent. AVS includes VMware licensing in the host price. EVS is BYOL, meaning customers stay in a commercial relationship with Broadcom for the VMware portion. This pushes organizations who want to stop paying Broadcom toward Azure, even when AWS is otherwise the preferred destination.
- Azure has a Windows licensing advantage. For Windows-heavy estates, Azure Hybrid Benefit makes the unit economics meaningfully better. This was true before Broadcom; it now matters more because the rest of the cost stack changed.
The pattern showing up in 2026 engagements is not a wholesale shift away from AWS, but a reweighting. AWS retains the lead for container-heavy, data-intensive, and Linux-dominated workloads. Azure picks up share specifically for the VMware exit and Windows server consolidation segment.
The detailed AWS vs Azure decision, applied in the opposite direction, is covered in the AWS to Azure migration guide.
4. Capex avoidance becomes a stronger argument than opex savings
For years, the cloud business case leaned on opex savings: lower TCO at scale, autoscaling, paying only for what you use. Customers who already had VMware hardware paid down often pushed back. Their argument: the sunk cost was real, the next refresh was not for three more years, and on-prem economics looked fine for now.
The licensing shift broke that argument. The next hardware refresh is still three years away, but the next software renewal is not. Customers who would have happily run their depreciated VMware estate until 2027 or 2028 now face a renewal cliff that they cannot wait out.
This has changed how cloud migration cases are written. The headline number is increasingly the avoided future capex and software cost, not the year-over-year opex savings. For finance leadership, the framing is closer to a refinancing decision than a transformation decision.
5. Greenfield workloads stop landing on VMware
The smallest but most directional shift: net new workloads are not being placed on VMware in 2026 unless there is a specific reason. A decade ago the default landing zone for a new application was a VMware VM on the existing cluster. In 2026 the default landing zones are managed Kubernetes (EKS, AKS, or GKE), managed VMs in the public cloud, or serverless.
This is happening even at organizations whose existing VMware estate is staying put. The strategic effect is that the VMware footprint stops growing organically, which over a few years means the renewal base shrinks even without active migration. The opposite of the 2010s default trajectory.
For the platform decision on the Kubernetes side, the EKS vs AKS vs GKE comparison covers the trade-offs that matter for enterprise landing zones.
How the response varies by enterprise segment
The strategic shifts above show up differently depending on the organization’s size, regulation, and existing cloud maturity.
Small to mid-market (under 200 VMware hosts)
Action is faster here. The renewal increase is large in absolute terms but small relative to the cost of an enterprise system integrator engagement, so internal teams or boutique partners can lead the work. The dominant pattern: skip the lift-and-shift step and go native to AWS or Azure directly.
Hypervisor alternatives like Proxmox and Nutanix also get more traction in this segment. The reduced enterprise feature set matters less when the operating environment is simpler.
Large enterprise (500 to 5,000 VMware hosts)
The phased pattern dominates here. Lift-and-shift to AVS first to escape the renewal cliff, then modernize app-by-app over 18 to 36 months. The total program runs 24 to 48 months. Larger organizations also tend to negotiate harder with Broadcom on the way out, sometimes securing 6 to 12 months of bridge pricing that buys planning time.
Internal politics matter at this scale. Application teams who own the workloads have to be aligned on the destination, and platform teams have to deliver the landing zone before app migrations can start. This is where the program management overhead becomes the bottleneck rather than the technology.
Regulated industries (financial services, healthcare, government)
These organizations move more slowly, both because compliance evidence has to be rebuilt on the new platform and because change risk is asymmetric: a botched migration in regulated environments costs more than the savings from a faster timeline.
The pattern: longer evaluations, more parallel proof-of-concept work, and a preference for the lift-and-shift step (AVS or sovereign-cloud equivalents) as a risk-reduction strategy. Native modernization happens, but on a slower clock and often only after the lift-and-shift is fully stable.
Compliance frameworks like HIPAA, PCI DSS, and FedRAMP add specific constraints that change which paths are viable. Healthcare organizations in particular tend to favor AVS over EVS because the Microsoft compliance posture is simpler to evidence than the AWS-plus-Broadcom split.
Public sector and sovereign cloud
A smaller segment but interesting. Public sector buyers facing both the Broadcom increase and growing sovereignty requirements are evaluating European and regional cloud providers (OVHcloud, Scaleway, G42 in the UAE, STC Cloud in Saudi Arabia) alongside the major hyperscalers. The non-VMware hypervisor alternatives are particularly relevant here, because some sovereign clouds offer Proxmox or KVM-based infrastructure that does not require any commercial hypervisor licensing.
How the math has changed for cloud-native vs lift-and-shift
A useful way to see the strategic shift is to look at how the relative attractiveness of the three main paths has changed since 2023.
| Path | Pre-Broadcom (2023) | Post-Broadcom (2026) |
|---|---|---|
| Stay on VMware | Default. Cheap, predictable, low risk | Most expensive option over 3 years for most estates |
| Lift-and-shift to cloud VMware (AVS, EVS) | Niche. Used for specific DR or burst cases | Mainstream. Fastest path out of renewal pressure |
| Native cloud modernization | Strategic. 5 to 7 year roadmap | Tactical. 18 to 36 month roadmap |
| Alternative hypervisor on-prem | Rarely considered | Actively evaluated in many engagements |
The shift in the last column is the strategic story. Paths that were considered exotic in 2023 are mainline in 2026, and paths that were the default are now the most expensive option.
For directional cost modeling of these paths against a specific estate, the VMware migration cost calculator compares 3-year TCO across Broadcom renewal, AVS, EVS, and native AWS or Azure.
What CIOs are getting wrong
Across the engagements happening this year, four patterns keep recurring where the strategic response is suboptimal.
Negotiating instead of planning
The temptation to negotiate the renewal down and defer the strategic decision is real. It usually backfires. Broadcom has limited incentive to discount durably, and a six-month renewal extension buys exactly six months, after which the same conversation happens with less leverage.
The organizations that come out of the negotiation in the strongest position are the ones who have a credible exit plan before the conversation starts. The credible plan, paradoxically, is what unlocks the negotiation. Without it, the renewal is non-negotiable.
Lift-and-shift as the destination, not the step
AVS and EVS are useful as transitional environments. They are expensive as permanent ones. The host-based pricing model means under-utilized capacity is fully paid for, and the architectural benefits of cloud-native (autoscaling, managed services, modern observability) are inaccessible inside the VMware boundary.
The CIOs who are landing in good positions are treating AVS or EVS as a 12 to 24 month hotel. The ones at risk are treating it as the new home.
Picking the target cloud by vendor relationship
Enterprise procurement relationships often nudge the decision toward whichever cloud has the strongest commercial team or the deepest existing relationship. This is not always wrong, but it is rarely the most important variable.
The variables that matter more for VMware migration: workload mix (Windows-heavy favors Azure, container and data-heavy favors AWS), existing identity and operations footprint, regional availability for regulated workloads, and the team’s experience and certifications. Picking the cloud first and forcing the workloads to fit is a common pattern and a common source of overruns.
Not budgeting for the modernization phase
Lift-and-shift to AVS or EVS is a project. Modernization is a program. Organizations that scope the migration as the project and treat modernization as something they will do later usually end up doing very little of it. The result is the worst of both worlds: the cost of the cloud landing zone plus the absence of the savings that justify it.
The CIOs avoiding this are explicitly budgeting and staffing the modernization work upfront, even when execution lands months after the initial lift-and-shift.
For more on the operational risks across cloud migrations broadly, common cloud migration challenges covers the patterns most likely to undermine these programs.
The strategic timeline question
The timing of the decision depends on where the renewal falls in the calendar. A simple framework:
- Renewal under 6 months out: The window for native modernization has closed. Negotiate a bridge, lift-and-shift to AVS or EVS, modernize after landing
- Renewal 6 to 12 months out: Both paths viable. The phased two-step pattern fits well here
- Renewal 12 to 18 months out: Native modernization is on the table for most estates. Decision should be made in the next 60 to 90 days to leave room for execution
- Renewal 18+ months out: Greenfield workloads should already be landing somewhere other than VMware. Existing estate migration can be planned for the next renewal cycle
The error pattern is to wait. Organizations that started planning at the 6-month mark are usually committed to lift-and-shift whether or not it is their best long-term option. Organizations that started at the 18-month mark have the option to choose.
For a structured pre-migration checklist that maps to these timelines, the enterprise cloud migration checklist covers the wider planning scope.
Questions worth asking before signing anything
Whether the answer is a renewal, a migration, or a phased plan, a small set of questions tends to surface the strategic implications that are easy to miss in the procurement conversation.
Procurement questions
- What is the full 3-year TCO of the renewal, including bundle components we are not using?
- What is the explicit price escalation clause for the second renewal?
- What contractual exit options exist if we begin migration during the term?
- What support and migration commitments are bundled vs separately priced?
Architecture questions
- Which of our workloads are tightly coupled to vSphere, vSAN, or NSX-specific features?
- Which workloads could run unchanged on any hypervisor or directly in a cloud VM?
- What dependencies exist on VMware-adjacent tooling (vRealize / Aria, Carbon Black, third-party automation)?
- What does our identity, networking, and storage estate look like outside of VMware?
Compliance and risk questions
- What evidence does our compliance posture currently depend on the VMware platform for?
- What does that evidence need to look like in the target environment?
- What is our auditor’s view of the proposed migration?
- What is our rollback position 6 months into a migration?
Capability questions
- Do we have the in-house skills for the target platform, or is partner support a structural dependency?
- What is the realistic timeline for the team to develop the target skills?
- How does our existing operations model translate to the new environment?
The depth of these answers usually correlates with how well the resulting decision will hold up under execution.
FAQ
Why did Broadcom raise VMware prices so much?
Broadcom acquired VMware in 2023 with a strategy to consolidate the customer base around bundled subscription products, prioritize the largest accounts, and accept the loss of customers who would not renew at the new pricing. The price increases reflect that strategy more than they reflect underlying cost changes.
Is VMware still a viable platform in 2026?
Yes. For organizations that are heavily invested, have specific operational dependencies on vSphere, and can absorb the higher pricing, VMware remains a working platform. The strategic question is not whether it works, but whether the cost-to-value ratio still justifies the dependency for any given enterprise.
What is the cheapest path off VMware?
Over a 3-year horizon, native cloud modernization is typically the lowest TCO, but it requires the highest upfront investment in migration work. Over a 12-month horizon, lift-and-shift to Azure VMware Solution or Amazon EVS is the lowest TCO path that escapes the immediate renewal. The cheapest path that an organization can actually execute is usually a phased combination of both.
How much does a typical VMware migration cost?
For a mid-sized estate of 500 to 1,000 hosts, the project cost for a phased lift-and-shift plus modernization is typically in the low to mid seven figures, including landing zone build, migration tooling, network re-architecture, and the application work. The cloud platform spend over 3 years is usually a multiple of the project cost. Both are typically less than 3 years of a 4x Broadcom renewal.
Should we consider Nutanix or Proxmox instead of cloud?
For organizations whose strategy is to stay on-prem but escape Broadcom pricing, Nutanix and Proxmox are credible alternatives. Nutanix offers the closer feature parity with VMware and the more enterprise-ready commercial support. Proxmox offers significantly lower software cost and a smaller operational footprint, with a feature gap that is closing but still real. The trade-off versus going to cloud is the same trade-off it has always been: control and predictability on-prem versus flexibility and scaling in cloud.
What does AVS or EVS lock-in look like?
Both are forms of vendor concentration. AVS concentrates dependency on Microsoft, including the bundled VMware license. EVS keeps the Broadcom dependency in place and adds AWS as the infrastructure provider. Neither is a permanent destination for cost-sensitive organizations. The lock-in becomes more concerning when AVS or EVS is treated as the final state rather than a bridge.
How is greenfield infrastructure being designed in 2026?
The default for net new workloads is managed Kubernetes (EKS, AKS, or GKE), managed cloud VMs for legacy applications, and managed databases (RDS, Azure SQL, Aurora) for state. VMware is rarely the default for new workloads in 2026. This is true even at organizations that maintain a substantial existing VMware estate.
Where infrastructure strategy goes from here
The VMware licensing shift is not a one-time event that resolves with a single decision. It is a structural change in the economics of enterprise infrastructure that will continue to reshape architecture, procurement, and operating model decisions through the rest of the decade.
For most enterprises, the practical implication is that the strategy work that was previously deferred to the next refresh cycle now has to happen on the next renewal cycle. The organizations adapting fastest are the ones that have treated this as a multi-year program, not a procurement event.
For a hands-on view of how the migration actually executes against this strategy, the VMware migration playbook covers the four real paths in detail and where they tend to break in production.
Need help shaping the strategy?
The VMware licensing shift puts most enterprises in front of an infrastructure decision they did not plan for. The right answer depends on the renewal timeline, workload mix, regulatory posture, and the realistic capacity of the internal team to execute against it.
Tasrie IT Services provides hands-on cloud migration services that cover:
- Strategic assessment of the VMware estate against renewal pressure, workload fit, and team capability
- Decision frameworks for choosing between AVS, EVS, native cloud, and on-premise hypervisor alternatives
- Migration delivery from landing zone design through wave-based cutover and post-migration optimization
- Modernization roadmaps that capture the long-term value of leaving VMware rather than stopping at lift-and-shift
The strategy is the harder part. The execution is the longer part. Both are easier with a team that has seen the patterns repeat across multiple engagements.