← InsightsCloud

Your VMware cloud provider can be sold from under you. Plan for it now

A VMware exit plan matters more than the migration date: IBM has sold contracts for hundreds of its VMware cloud customers to 11:11 Systems. Check your contract and your options now.

What happened

On 8 October 2026, The Register reported that IBM has sold contracts for "hundreds" of its cloud-hosted VMware customers to 11:11 Systems, a Broadcom partner. IBM did not respond to the reporter's questions.

The background is Broadcom's reduction of its cloud partner programme. According to the report, Broadcom kept only partners focused on VMware Cloud Foundation (VCF). Partners it excluded lost the right to resell its software in October 2025 and may not run VMware-powered clouds after March 2027. IBM stopped selling new cloud VMware services in 2025.

11:11's chief executive, Brett Diamond, told The Register that the contracts cover customers on VCF versions 7 and 8, which he described as "ripe for migration". 11:11 plans to move them to VCF 9, because, he said, VMware ends support for VCF 8 next year. This is 11:11's eighth acquisition of a VMware service provider. The article speculates that IBM's Red Hat OpenShift business may be one reason IBM was excluded, since it competes with VMware.

The article also records a wider dispute. Broadcom says that services firms not dedicated to VCF will not meet customer needs. The lobby group CISPE argues that Broadcom has improperly concentrated the market and is pursuing an antitrust case in EU courts.

Why it matters

Your supplier can change without your say. If your VMware workloads run at a hosting provider, the company you chose may no longer exist in that form by the time your contract ends. The report describes customers being moved to another provider as part of a commercial deal between two companies.

A forced move comes with a forced upgrade. On the account of 11:11's chief executive, the customers acquired are on older VCF versions and will be moved to VCF 9. A migration that arrives on the new owner's schedule is rarely the one you would have designed.

Dates set outside your organisation now drive your plan. The March 2027 limit on partners running VMware-powered clouds and the end of support for VCF 8 are not your deadlines, but they define your options. The sooner you know which of your workloads they touch, the more choice you keep.

Exit planning is also a sovereignty and resilience question. A recent announcement shows how seriously the largest cloud providers now take independence from their own parent companies. On 28 September, AWS announced that it would run its European Sovereign Cloud for several hours on Saturday 24 October without a connection to the AWS global network, with an all-EU-resident team, to show it can operate independently. We have not seen the results, and AWS describes the exercise as aligned with European sovereignty objectives, not as certification. The point for buyers is that proof of operational independence is becoming something to ask suppliers for, whichever platform you use.

Our view is that most organisations have a migration plan that depends on a single provider and a single timeline. The better plan has options.

What we would do

  1. Find out where your VMware actually runs. List every workload on a hosting partner, on your own hardware and on a hyperscaler's VMware service, with the VCF version, the provider and the contract end date.
  2. Read the contract for what happens on a sale. Check assignment, change of control, notice, data location and termination terms. Ask your lawyers to confirm your position, and ask the provider in writing what its plans are.
  3. Sort workloads by their dependence on the hypervisor. Some can move to a managed container or platform service with little change, some need re-hosting, and some should stay put. This sorting, not the platform choice, sets your cost and risk.
  4. Cost three routes, not one. Compare staying on VMware with a chosen partner, moving to a hyperscaler, and re-platforming selected applications. Include licence terms, the cost of support at older versions, and the cost of doing nothing.
  5. Keep an exit you have tested. Make sure you can extract your data and configuration in a usable form, and rehearse restoring one workload elsewhere. An untested exit is a hope.
  6. Ask suppliers for evidence of independence. Where residency or sovereignty matters, ask providers how they operate if a parent company, partner or network link changes, and what proof they can share.

The aim is not to leave VMware in a hurry. It is to make sure the next change of ownership, whoever's it is, is an inconvenience and not an emergency.

Questions we are asked

Do we have to accept the new provider if our contract is sold?

That depends entirely on your contract, which is why it is worth reading now. Look at assignment and change-of-control clauses, notice periods, service levels, data location commitments and termination rights. We are not giving legal advice, so take it from your lawyers, but you should know your position before a notice arrives rather than after.

Is moving to a hyperscaler the only alternative?

No. The realistic options include staying on VMware with a provider or on your own estate, moving workloads to a hyperscaler, using a Kubernetes-based platform for suitable applications, or a mix. The right answer varies by workload. Sorting applications by how much they depend on the hypervisor is usually the first useful step.

Why does this matter if our provider has not said anything?

Broadcom's partner changes mean provider ownership can change quickly and with limited choice for the customer. A provider that is calm today may be acquired, restricted or withdraw from the market. A tested exit plan and a current inventory cost little to prepare and are very expensive to improvise under a deadline.

Sources

  1. IBM offloads 'hundreds' of its cloudy VMware customers — The Register
  2. AWS European Sovereign Cloud: Demonstrating an independent operation — AWS Security Blog