The License You Already Bought

💸The Squeeze#

Paid-for software, re-billed as subscription

Broadcom sought what Tesco's filing calls excessive prices for software Tesco had already paid for: a ~175% VMware hike and ~350% mainframe upcharge, and a $23.5M offer for a single year of Cloud Foundation 9.0 plus mainframe support. The pattern is the lock-in playbook, the cost of staying is set after the customer is already embedded.

📅Renewal Timing#

The calendar decides the budget

Each subscription renewal is a fork: sign and you commit another term on the platform you are trying to leave; refuse and you race a hard cutover date. Migration timelines must be planned backward from renewal dates, not forward from project kickoff.

🚪Vendor Lock-In as a Market#

Exit pain becomes someone's TAM

When the cost of staying spikes, the cost of leaving becomes a business. Migration tooling, assessment services, and alternative hypervisor vendors all sell against the same squeeze, and HPE is offering VMware refugees a free year of virtualization to pull them across.

📈What the Industry Learned#

150-1000% is a structural break

Broadcom's repricing drove broad 150-1000% hikes and aggressive subscription bundling across the VMware base. The lesson buyers are internalizing: perpetual licensing and acquirable vendors are a latent risk that only prices in after an ownership change.

⚖️£100M and the Exit Clock#

Litigate while you migrate

Tesco filed a ~£100M UK High Court claim against Broadcom, VMware, and reseller Computacenter, and set an end-2027 deadline to move 40,000 server workloads. The lawsuit and the migration run in parallel: legal leverage buys time, but the renewal calendar, not the court, sets the real schedule.

🛠️The Engineering Reality#

Lift-and-shift is rarely just a shift

Moving tens of thousands of workloads is a 6-24 month program, not a weekend cutover. Hypervisor swap is the easy part; the hard part is matching vCenter's HA/DRS/SRM behavior, storage integration, and operational tooling that teams built their runbooks around.

🧩Proxmox VE#

Open-source, mid-market default

For 50-200 VMs with mixed workloads, Proxmox VE is the community consensus replacement: core compute, storage, and HA parity with no license fee and optional paid support. It preserves VM-based workflows without re-architecting applications.

🏢Nutanix AHV#

Enterprise HA/DRS-class swap

At 200+ VMs with SRM/DRS complexity, Nutanix AHV plus Nutanix Move is the realistic enterprise path, Prism Central rivals vCenter for one-click management, and Move handles the conversion. The trade is a different platform's operational model, not zero migration.

☁️OpenStack & Hyper-V#

When scale or existing stack decides

OpenStack fits large multi-tenant private clouds with hard storage requirements and in-house ops depth. Hyper-V suits shops already standardized on Windows Server. Both trade VMware's polish for control or incumbency.

🎚️Decision Tiers#

VM count and timing pick the target

The destination is a function of scale and clock, not preference: small/mid estates lean Proxmox, large HA-heavy estates lean Nutanix or a managed exit, and the renewal deadline determines how much re-architecture you can afford versus a straight lift-and-shift.