Never thought I'd be able to say this, but Broadcom has out-CA'd Computer Associates.
For those not familiar with CA, their model was to buy up failed enterprise software companies and milk the s@*t out of the licensing and support contracts, while spending next-to-nothing on development. Broadcom follows much the same model with steroids and protein powder.
CA failed, only after years of profitable rent-seeking. The same may happen to Broadcom, but it will take years. In the meantime the companies who depend on the software they control will pay a heavy price.
CA was horrible. I worked there but your claims are a little too strong. They bought up and coming products within rising sectors, "over paying" knowing they could cash in. They then introduced those products into their large portfolio of large enterprise clients (who trusted them deeply) giving the purchased products increased market share and real associated profits. They were the trusted technical friend for non technical companies. They did after initial investment then largely and entirely killing innovation on and investment in the products. It was the place that products went to die.
[edit: perhaps it got darker and more cynical after I left. I wouldn't be terribly surprised. I left in 2013, post Enron style accounting fraud, pre sale to Broadcom]
I do suspect they'll settle but I doubt it will result in a new contract. They've already made significant investments migrating off VMWare and none of those will snap back with a new contract. Tesco was also cut off from updates and support, and so had to purchase 3rd party support, so they have real damages.
Making on premise hosting more costly changes the calculus making cloud migrations easier to sell to the board. Broadcom's revenue is literally a slice of hyperscaler capex and that capex is enormous. Combined AI infrastructure spending across the major players runs well past $600 billion annually. This aligns with the federal government's need to have data custody and supports whatever the Communications Assistance for Law Enforcement Act (CALEA) is called now. They are geniuses.
A supermarket with over 3000 (5000 depending where I search) stores probably doesn't want everything in the cloud. They will want every shop to be at least mostly-capable of operating without an internet connection.
You're not doing business with broadcom there. Though most people working at the raspberry foundation are broadcom people. Most from the time before broadcom turned full evil though.
I agree about boring tech. They're not a tech company but they're not low or anti tech though
Tesco worked closely with a data startup to create the Clubcard and that startup is now a big company owned by Tesco (Dunnhumby, £350m, 2,000 employees)
They were the first supermarket to do online grocery delivery
They had Nokia apps for groceries on 2009, iPhone in 2010
Eh, it's still the least hassle at scale for your own data centers and VCF 9 isn't exactly low tech - it does things no one else really can do to the same level of quality (DRS, HCX, NSX, VSAN all come to mind) , the question is, is it worth the money? Unless you're over 10,000 VMs, probably not.
OpenShift is okay but has quirks unless you have a major Kubernetes staff; Proxmox is good for most but I wouldn't use at massive scale. Azure Stack/ Hyper-V can work but has its own quirks.
They're not the only ones pissed off with VMWare [1].
What I don't understand is what prevents them to shop around for alternatives. What is VMWare's very good moat that prevents the competition from invading their castle?
2. Tesco had a good deal they paid for ahead of time, with the full support cycle detailed in the contract that Broadcom is refusing to honor.
3. Broadcom is arguing that since Tesco is migrating off, they can't sue for damages. Tesco would probably counter argue that the migration is due to Broadcom not honoring the original contract.
20 years ago anyone and everyone had datacenters full of VMware. Now most of it is stuff that a company hasn't bothered to move to cloud yet. There's a few things it's still the best choice for but those are niche things. Like stuff that absolutely needs to be self hosted for privacy or security. The time of VMware being the default is long gone.
Moving from a mainstream to a niche product means a market that's shrinking. No growth potential. No potential for new competitors to start up. Proxmox is good but it's not exactly enterprise.
This is why Broadcom bought it and sucks the most value out of it before it completely disappears.
> There's a few things it's still the best choice for but those are niche things.
That just isn't true. The reality is that being in the cloud is not beneficial for most businesses. They do it because it's trendy, not because it's the best solution. In recent years we've seen more and more recognition of that fact, and it has driven people to host stuff on prem.
Well that's another story, which I totally agree with by the way. I'm not a proponent of cloud at a all for 24/7 workloads.
However the fact is that most companies do this anyhow and they feel like it's the best solution. Onprem vmware is just a shrinking market. I don't really see that expanding again until it hits rockbottom and of course broadcom's behaviour isn't helping.
because there are no real alternatives. Everything that could be considered an alternative has drawbacks or things that are missing. Proxmox comes close but doesn't offer proper enterprise support contracts, so you'd be stuck with a 3rd party.
Then there's training. you can't easily switch your admins and service desk techs to a different product. That alone takes months, of not years, and costs a lot. Rewrite all processes, etc.
Then there's 3rd party integration. Since VMware was basically the "default", most 3rd party products offered turnkey integration into VMware, and VMware only. Think backup applications or security etc. You don't switch backup vendors easily (for the same reasons - training, features, ...) and if you do consider it, it adds to the cost
This is why, for many companies that don't have 50-100 people or more in their IT department, it's more expensive to switch away from VMware so they grudgingly pay, while trying to move as much workload away from it as possible.
Never thought I'd be able to say this, but Broadcom has out-CA'd Computer Associates.
For those not familiar with CA, their model was to buy up failed enterprise software companies and milk the s@*t out of the licensing and support contracts, while spending next-to-nothing on development. Broadcom follows much the same model with steroids and protein powder.
CA failed, only after years of profitable rent-seeking. The same may happen to Broadcom, but it will take years. In the meantime the companies who depend on the software they control will pay a heavy price.
CA was horrible. I worked there but your claims are a little too strong. They bought up and coming products within rising sectors, "over paying" knowing they could cash in. They then introduced those products into their large portfolio of large enterprise clients (who trusted them deeply) giving the purchased products increased market share and real associated profits. They were the trusted technical friend for non technical companies. They did after initial investment then largely and entirely killing innovation on and investment in the products. It was the place that products went to die.
[edit: perhaps it got darker and more cynical after I left. I wouldn't be terribly surprised. I left in 2013, post Enron style accounting fraud, pre sale to Broadcom]
Feel like it's burying the lede a bit to not mention that Broadcom acquired Computer Associates and said "hold my beer".
Facebook and good should be at the top of that list for selling scams to users
"In 2018, Computer Associates was acquired by Broadcom Inc., a semiconductor manufacturer, for nearly $19 billion."
The behaviour was acquired too ....
Negotiating tactic. Never really makes it to court.
I do suspect they'll settle but I doubt it will result in a new contract. They've already made significant investments migrating off VMWare and none of those will snap back with a new contract. Tesco was also cut off from updates and support, and so had to purchase 3rd party support, so they have real damages.
Making on premise hosting more costly changes the calculus making cloud migrations easier to sell to the board. Broadcom's revenue is literally a slice of hyperscaler capex and that capex is enormous. Combined AI infrastructure spending across the major players runs well past $600 billion annually. This aligns with the federal government's need to have data custody and supports whatever the Communications Assistance for Law Enforcement Act (CALEA) is called now. They are geniuses.
A supermarket with over 3000 (5000 depending where I search) stores probably doesn't want everything in the cloud. They will want every shop to be at least mostly-capable of operating without an internet connection.
VMware really damaged their business longterm, as anyone that can transition off, is doing so, and anyone else is planning on it.
Broadcom doesn't care. They just want to milk VMware for all it's worth to goose their short-term profits.
Yeah this was clearly willingly and knowingly done.
They did the same rugpull to PLX customers. This is Broadcoms business model. Never ever do business with Broadcom.
Difficult for me because I really like Raspberry Pis.
You're not doing business with broadcom there. Though most people working at the raspberry foundation are broadcom people. Most from the time before broadcom turned full evil though.
They explained it to their investors when they announced the purchase
Having VMware in 2026 is a sign of low tech.
Supermarkets probably should use boring tech. They aren't tech companies.
I agree about boring tech. They're not a tech company but they're not low or anti tech though
Tesco worked closely with a data startup to create the Clubcard and that startup is now a big company owned by Tesco (Dunnhumby, £350m, 2,000 employees)
They were the first supermarket to do online grocery delivery
They had Nokia apps for groceries on 2009, iPhone in 2010
Huge supply chain automation
Tesco Labs etc
Eh, it's still the least hassle at scale for your own data centers and VCF 9 isn't exactly low tech - it does things no one else really can do to the same level of quality (DRS, HCX, NSX, VSAN all come to mind) , the question is, is it worth the money? Unless you're over 10,000 VMs, probably not.
OpenShift is okay but has quirks unless you have a major Kubernetes staff; Proxmox is good for most but I wouldn't use at massive scale. Azure Stack/ Hyper-V can work but has its own quirks.
Compared to something like proxmox? Sure.
It takes a long time to change corporate mindset.
Compared to something like openstack.
Comments like this are a sign of someone living in a bubble.
Other than proxmox, what alternative do you have for "high tech"?
(2025) OP.
The current report and discussion this week:
Tesco moving 40k server workloads off VMware amid Broadcom's abusive conduct
https://news.ycombinator.com/item?id=48576838
They're not the only ones pissed off with VMWare [1].
What I don't understand is what prevents them to shop around for alternatives. What is VMWare's very good moat that prevents the competition from invading their castle?
https://www.theregister.com/software/2026/03/24/half-of-vmwa...
1. Tesco is migrating off
2. Tesco had a good deal they paid for ahead of time, with the full support cycle detailed in the contract that Broadcom is refusing to honor.
3. Broadcom is arguing that since Tesco is migrating off, they can't sue for damages. Tesco would probably counter argue that the migration is due to Broadcom not honoring the original contract.
Because there's no business potential.
20 years ago anyone and everyone had datacenters full of VMware. Now most of it is stuff that a company hasn't bothered to move to cloud yet. There's a few things it's still the best choice for but those are niche things. Like stuff that absolutely needs to be self hosted for privacy or security. The time of VMware being the default is long gone.
Moving from a mainstream to a niche product means a market that's shrinking. No growth potential. No potential for new competitors to start up. Proxmox is good but it's not exactly enterprise.
This is why Broadcom bought it and sucks the most value out of it before it completely disappears.
> There's a few things it's still the best choice for but those are niche things.
That just isn't true. The reality is that being in the cloud is not beneficial for most businesses. They do it because it's trendy, not because it's the best solution. In recent years we've seen more and more recognition of that fact, and it has driven people to host stuff on prem.
Well that's another story, which I totally agree with by the way. I'm not a proponent of cloud at a all for 24/7 workloads.
However the fact is that most companies do this anyhow and they feel like it's the best solution. Onprem vmware is just a shrinking market. I don't really see that expanding again until it hits rockbottom and of course broadcom's behaviour isn't helping.
Big ships turn slowly is all. It’s no easy feat to move away from something like that.
because there are no real alternatives. Everything that could be considered an alternative has drawbacks or things that are missing. Proxmox comes close but doesn't offer proper enterprise support contracts, so you'd be stuck with a 3rd party.
Then there's training. you can't easily switch your admins and service desk techs to a different product. That alone takes months, of not years, and costs a lot. Rewrite all processes, etc.
Then there's 3rd party integration. Since VMware was basically the "default", most 3rd party products offered turnkey integration into VMware, and VMware only. Think backup applications or security etc. You don't switch backup vendors easily (for the same reasons - training, features, ...) and if you do consider it, it adds to the cost
This is why, for many companies that don't have 50-100 people or more in their IT department, it's more expensive to switch away from VMware so they grudgingly pay, while trying to move as much workload away from it as possible.
> Proxmox comes close but doesn't offer proper enterprise support contracts
... but then again, neither does VMware, apparently.
dupe
https://news.ycombinator.com/item?id=48576838