VCF Switching Cost Is the License Question
Broadcom sells VMware Cloud Foundation as a full-stack private cloud: compute, storage, networking, security, and management in one suite. There is no one-for-one replacement on a rival hypervisor. Bills after a perpetual-to-subscription conversion have, for many estates, moved several times over. Most of you will shrink rather than exit. Gartner has forecast that 35 percent of VMware workloads will sit on a different platform by 2028, which still leaves most of the estate where it is.
That’s the problem.
The parent pack is the VMware Broadcom stay vs exit page: perpetual versus VCF, the support path, portable VCF on AVS, and third-party cover.
This is not a hypervisor bake-off. The missing one-for-one VCF replacement is a commercial bet against your switching cost. The hike walks in as a number on this year’s bill. The exit walks in as more platforms, more hardware, and the applications you cannot unwind. If you cannot price those two things against the estate you already run (this week), the vendor has already framed the meeting.
What Broadcom actually sells, and what a swap would cost
VCF 9 is documented as a full-stack IaaS platform. It is not a la carte vSphere with optional add-ons. That is the commercial object. Previously you could pick parts of the stack. Broadcom now sells the complete suite.
Treat any “typical” percentage hike as typical, not as your estate. It does not tell your CFO what your quote is. Broadcom’s answer is that the full-stack private clouds VCF makes possible are efficient enough that the suite pays for itself. That is an efficiency claim, not a license position, until you recast it against this deployment.
A migration is a significant engineering project. It needs an assessment of every application. Not every workload is certified on a non-VMware hypervisor. Some vendors now ship cloud-native versions: an easier on-ramp to containers. Explore those. Do not make architectural decisions that close the door on leaving.
Gartner’s 35 percent figure is a workload forecast, not a census of customers leaving. Thirty-five percent of workloads elsewhere still leaves most of the estate on VMware. Most organizations will minimize the footprint rather than eliminate it. Application dependencies are not easy or economical to unwind, so a reduction means multiple replacements and more infrastructure to manage. Rival hypervisors rarely match VMware’s VM density, so moving often means buying more hardware.
Public cloud and HCI are the usual destinations. Public clouds will often welcome a large VM fleet with migration services. HCI tooling quality varies. OpenStack is too big for a typical IT shop to run as an exit plan. None of those sentences is a license position.
The hike is not your switching cost
Look at your VMware file. A product row and a renewal date do not tell you whether the quote walking in is a like-for-like refresh or a VCF-only conversion.
If your reports still treat “leave VMware” as a destination list price, you are reading a reporting layer. You have a hypervisor count. You do not have a commercial outcome.
This is not a criticism of the teams building those reports. Traditional SAM tools were built to prove completeness, not to price a VCF bill against a partial move this week. An analysis that arrives after the vendor has framed stay-or-leave is a document. One that arrives during the meeting is a decision input.
Picture your procurement director, sixty days from a VMware conversation. The CIO has the line: there is no one-for-one replacement. Finance has a several-times-over hike. Infrastructure has a slide that says public cloud or HCI. None of those is a license position.
The stay number is the VCF quote on the estate you run. The leave number is not a rival’s list price. Switching cost is the residual VMware you cannot unwind, plus extra platforms, extra hardware, and the engineering to assess every application. If a workload is not certified elsewhere, that line is still a VMware line.
Do not let the vendor frame the meeting as “do you want a modern private cloud” or “do you want to get off Broadcom.” The question is what the hike costs on this estate versus what a shrink or a partial exit costs once you price the complexity.
Most of you will shrink. That is still a license event.
A plan to move a slice does not retire the VCF conversation on what stays. A modernization program does not recast the bill on the hypervisor those applications still sit on. If you cannot say, this week, which workloads can move, what they cost to move, and what VCF still costs on the remainder, you are walking in with a slide. Broadcom is walking in with a suite.
If that feels familiar, the problem is not your team’s maturity. A commercial bet is being answered with a destination rumor. That shrink-versus-exit fork is already in UK and EU files: Tesco’s lawsuit is the customer who would not stay; Siemens is the inventory reminder if you do; CISPE’s challenge to Broadcom’s VMware deal is the European overlay.
The number to have before you pick a path
Do this before the next VMware conversation. Produce a decision pack, not a migration program.
| Path | What the market is saying | The number you lock first |
|---|---|---|
| Stay on VCF | No one-for-one replacement. Suite-only conversion. Broadcom says the stack pays for itself. | Current entitlement versus the VCF-only quote on this estate. A “typical” hike is not your number. |
| Shrink | Most will minimize, not eliminate. Dependencies keep part of the estate. 35 percent of workloads elsewhere by 2028 is a workload forecast. | Named workloads that can move, certified. Residual VCF cost on what stays. Switching cost on the slice: extra platforms, extra hardware. |
| Exit | Few can quit entirely. Multiple replacements. More hardware. A fleet-wide assessment. | Applications you cannot unwind. Certification gaps. Hardware add for density loss. Residual Broadcom cost through cutover. |
Can a non-specialist brief the CIO in fifteen minutes from the pack you have today? If the answer is no, you do not have a license position. You have a conference quote.
Your job is not to decide whether a rival hypervisor is less capable. It is to walk in with the hike priced against the switching cost on this estate.
The decision layer, not another inventory
You already have discovery data and a renewal date. The gap is not another inventory. The gap is turning that estate into a decision: what VCF costs against what you operate, what a shrink still leaves on the bill, and whether an exit is a priced switching cost or a slide. LICENSEWARE sits on the inventory and ITSM tools you already run. It is not a rip-and-replace SAM suite. It is a decision layer: what matters, why it matters now, what should happen next. If you stay, audit defense and host topology from IFMP and RVTools / vCenter on data sources are the numbered file, not a destination rumor; Citrix’s post-Broadcom hypervisor U-turn is one more proof there is no one-for-one swap.
If you are heading into a VMware renewal or a Cloud Foundation conversation and your current tools still need three weeks to turn a stay-versus-shrink plan into a priced position, book an Audit Readiness Review. You can also start on the free plan and run the analysis on your own data.
The question to walk in with
Do not let the vendor frame this as “there is no replacement, here is Cloud Foundation.” The right question is: what are we actually running, what does VCF cost against that estate, which workloads can move without a certification hole, and what does the residual plus the extra platforms cost against the hike walking in. That is a data question, not a sales question.
The bet that you cannot move may be a good one. Do not make it easy. The vendor will walk in with a number. The only question is whether you have yours first: current, defensible, and tied to the contract in front of you.
