Tesco's engineering leadership publicly characterized Broadcom's licensing behavior as 'abusive conduct' — unusually blunt language for a FTSE 100 company. The 3x-10x renewal hikes, killed perpetual licenses, and forced bundle SKUs made the cost and risk of migrating 40,000 workloads preferable to staying on vSphere.
Broadcom's Hock Tan playbook assumes migration cost always exceeds the cost of staying. Tesco — running mission-critical retail workloads on aging vSphere clusters — just proved the math flips at sufficient pain, sending a signal to every vSphere shop facing a 2026 renewal that exit is now viable rather than theoretical.
The HN discussion repeatedly surfaces Proxmox, KVM-based platforms, and Nutanix AHV as the names Tesco and similar shops are landing on. The community consensus is that the technical portability gap that historically protected VMware has narrowed enough that a 40,000-workload migration is plausibly executable.
Tesco — the UK's largest grocer, ~£69B in annual revenue, ~330,000 employees — is moving roughly 40,000 server workloads off VMware. The Computer Weekly scoop, picked up by Ars Technica and pushed to 336 points on Hacker News, quotes Tesco's distributed systems engineering lead describing Broadcom's post-acquisition licensing behavior in unusually blunt terms for a FTSE 100: "abusive conduct."
The trigger isn't technical; it's commercial. When Broadcom closed its $69B VMware acquisition in late 2023, it killed perpetual licenses, collapsed the SKU catalog into a handful of bundles (VCF, VVF), and pushed minimum-core commitments that hit existing customers with 3x–10x bill increases at renewal. AT&T sued. Computershare went public with a $20M-plus hike. The European Cloud Competition Observatory has been collecting affidavits. Tesco is now the largest named enterprise to actually pull the trigger on migration rather than just complain.
The replacement stack reportedly leans on Nutanix AHV and open-source hypervisors — Proxmox and KVM-based platforms are the names that keep surfacing in the HN thread. Tesco hasn't published a target completion date, but 40,000 workloads is a multi-year program even with strong tooling. The story matters less for what Tesco is moving *to* than for the signal it sends to every other vSphere shop staring at a 2026 renewal quote.
Broadcom's strategy under Hock Tan has been consistent across CA, Symantec, and now VMware: identify the top ~600 customers who can't credibly leave, raise prices until they squeal, and let the long tail churn. The playbook assumes the cost and risk of migration is always higher than the cost of staying. Tesco just demonstrated that, at sufficient pain, the math flips even for a retailer running mission-critical workloads on vSphere clusters older than some of its engineers.
The interesting technical wrinkle is *what's actually portable now*. Five years ago, an enterprise VMware exit meant rebuilding NSX networking, replacing vSAN, finding a new VDI stack, and re-certifying every ISV. In 2026, the gap has narrowed: Nutanix AHV ships with Flow (microsegmentation) and a competent storage layer, OpenShift Virtualization runs KubeVirt under a Red Hat support umbrella that satisfies most procurement teams, and Proxmox VE 8.x has finally crossed the line for stateful workloads at scale. None of these are drop-in for the most exotic VMware features — but most enterprises were only using maybe 20% of vSphere's surface area anyway.
The community reaction on HN is worth reading because it's not gloating, it's logistics. Operators are trading notes on live migration paths (Veeam's instant recovery to AHV, the Move tool, OpenShift's MTV operator), on which workloads break (anything tightly coupled to vRealize, NSX-T overlays, or vSphere with Tanzu), and on the political dynamics of getting application teams to retest. The recurring theme: the technology is the easy part. The hard part is the spreadsheet that explains to the board why you signed a 5-year EA in 2022 and are now eating a write-off to escape.
Broadcom's response so far has been to do nothing publicly and lean on the contracts. That works until it doesn't. A few thousand SMB customers leaving for Proxmox doesn't move CFO Kirsten Spears's quarterly numbers. Tesco at 40,000 workloads is a different category of signal — it's the kind of reference customer competitors put in a slide deck and walk into every Global 2000 procurement meeting with.
If you run VMware in production, three concrete actions for the next 90 days:
Get a real number on portability. Inventory your VMs by feature dependency: how many use NSX overlays vs. straight VLAN, how many depend on vSAN-specific features vs. external storage, how many use vRealize Automation vs. Terraform/Ansible. The workloads that are just "a VM with a disk and a NIC" — typically 60–80% of an estate — are portable to any modern hypervisor in a weekend. Knowing that ratio is your renewal leverage.
Price out two alternatives, not one. The mistake Broadcom is counting on is that customers will treat migration as binary — stay or go. The smarter play is a credible bid from Nutanix *and* a credible bid from Red Hat OpenShift Virtualization (or Proxmox + a support partner), used as price discovery in your VCF renewal. Multiple shops on HN report 30–50% Broadcom discounts materializing the moment a signed Nutanix POC lands on the account exec's desk.
Watch the ISV certification matrix. The unsung blocker on a VMware exit isn't the hypervisor — it's the third-party software vendor whose support contract says "vSphere only." Oracle, SAP, and the EHR vendors are the usual suspects. The Tesco migration will, over the next 12 months, force a wave of ISV recertification on AHV and KubeVirt that benefits every other enterprise that follows. Free-ride that work.
The 2024–2025 VMware story was rate hikes and lawsuits. The 2026 story is reference migrations at scale, and Tesco just rang the bell. Watch for two follow-ons: which European bank or telco names itself next (the Computer Weekly piece hints at "several others" in the same FTSE 100 cohort), and whether Broadcom blinks on renewal pricing for tier-2 accounts to slow the bleeding. If neither happens by year-end, the Nutanix and Red Hat sales teams are going to have a very loud Q4.
”Tesco, a retail conglomerate headquartered in the United Kingdom”For any non Uk people, it’s the largest supermarket in Uk. Combination of large stores and smaller high street convenience stores.(2nd largest was owned by Walmart who sold it recently to private equity and so now it’s saddled with de
I worked in software acquisitions for a large organization and it was really eye opening to see how insane some of these companies are when it comes to pricing customers out. I always wondered - what is the motive? They make pricing structure changes that aren't even considerable for any organi
If Tesco needs character witnesses that Broadcom has done this to many other customers, I think they’ll find plenty of willing participants.Broadcom’s marketing for Proxmox is extremely effective.
> Tesco is also dealing with migration challenges related to data security because its new, unnamed virtualization software is incompatible with the Veeam and Zerto products it uses.What is a VMware alternative, that isn't compatible with backup software? I'm guessing it's not nuta
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
I’m surprised by the comments here.The Broadcom business model (outside the chip business) had been pretty well known, and they don’t really hide it.They are tech bottom feeders. They find large businesses with a decent moat and free cash flow but are in long term decline (and wasting cash trying to