Broadcom 'bulldozes' VMware cloud partners as March deadline looms

Broadcom this week brought the hammer down on the Advantage Partner Program for VMware Cloud Service Providers (VCSPs) – and the clock is now ticking for any third parties working to close sales. For customers, this potentially represents a massive supplier shake-up, with sources claiming that hundreds of CSPs across Europe …

[Groo The Wanderer - A Canuck]

When you only run a couple of VMs on a Linux box, it's VirtualBox for the win! To hell with Broadcom... they screwed me out of my "perpetual license" for VMWare Workstation. There is no way under any circumstances I will ever use one of their products again - even at a client site. Get someone else to do it...

[Anonymous Coward]

Broadcom made VMware Workstation free of charge for home and business users - one of the few good decisions there regarding the whole VMware portfolio.

I don't understand your problem.

[mattaw2001]

Careful, I got contradicted and down voted for saying this about VMWare Workstation a few months ago.

I even posted the link to VMWare's announcement that it was free for all uses, commercial or personal. I guess the professional pain suffered was overriding the facts for folks.

[Groo The Wanderer - A Canuck]

Exactly. I paid for a perpetual license and they abandoned that deal as soon as they took over. I don't give a rat's fat ass what they did later - they screwed me over! To HELL with them!

They forced me to find an alternative with no notice - I am NOT going back to them after that. They had me as a "perpetual customer" -- they're the ones who threw that relationship away, not me. I feel absolutely zero guilt for hating the company that screwed me out of what I paid for!

[Gene Cash]

Yes... basically "pray I don't alter the deal further"

[Anonymous Coward]

Where does it say your existing license isn't valid anymore ?

[TonyJ]

There were some examples posted online last year, of cease and desist letters being sent from Broadcom to customers, telling them that they could no longer use their perpetual licenses...

[Sandtitz]

Wrong.

The perpetual licenses are valid and users can still use them in perpetuum.

However, the users are NOT licensed for any updates after their support cut off date. So when your support contract ended, you are not allowed to install any updates dated after that date

VMware (pre-Broadcom) didn't enforce this - vCenter would happily download later versions of hypervisors (of same major version) or you could download the latest ISO and use it with your license key. It's not unheard of IT personnel just updating software accidentally without knowing their support service had ended, because the licenses are handled by other personnel.

In these cases Broadcom noticed the downloads and sent those letters telling users to stop using the updated software and to revert back to versions they were entitled to use.

This hardly is rare in enterprise products - firewalls, storage and other appliances typically restrict the download and installation of software unless your entitlement is valid.

[TonyJ]

Wrong, eh?

https://arstechnica.com/gadgets/2025/05/broadcom-sends-cease-and-desist-letters-to-subscription-less-vmware-users/

And if you read through it, because at first, and on the face of it, it doesn't start off too unreasonable-sounding: i.e. stop getting updates outside of your support contract, as you say, you will see this:

"...Some customers of Members IT Group, a managed services provider (MSP) in Canada, have received this letter, despite not receiving VMware updates since their support contracts expired, CTO Dean Colpitts told Ars. One customer, he said, received a letter six days after their support contract expired..."

So notwithstanding they were seemingly allowing out-of-support systems to update themselves, they also targeted owners of systems who hadn't done this.

And I would argue given the complexities of large companies, where, as you say the license renewals are handled by different departments, it's on the supplier to block such downloads in the first place once a support license expires. Others manage it fine.

[Sandtitz]

"it's on the supplier to block such downloads in the first place once a support license expires. Others manage it fine."

Which is why Broadcom stopped updates year ago. You need to add a customer specific token to the download URLs.

EMC/Dell didn't block VMware updates.

I think some people really thought their perpetual license not only includes software updates but also expected their current version to be under maintenance until heat death of the universe, or to have entitlement to every future version as well.

[retiredFool]

Question

I don't use any virtualization, so I don't keep up. I do see the headlines about this fiasco though. So the question is, can bcom revoke your free access to this workstation product in the future? If they can, I'd assume it is a matter of when not if given their prior behavior.

[Anonymous Coward]

Re: Question

There is no product activation or even installation keys. It works in machines without any internet access.

[retiredFool]

Re: Question

There could still be a time bomb embedded in it. Reason of course is "we want to make sure you update every n years"

[Anonymous Coward]

Re: Question

"There could still be a time bomb embedded in it."

You are just spreading FUD there. Too easy to move clock forwards and check for this.

Broadcom would lose a sizable chunk of their business rather immediately if they started to plant hidden time bombs in their code.

Username checks out.

[Roland6]

After discontinuing the previous perpetual licences.

If memory serves me correctly there was a gap of a few weeks between the withdrawal of the old product and the reintroduction of the free for personal/home use version.

Hence both of you are right.

[Jou (Mxyzptlk)]

While this is true, the actual need is for clustered high availability. The workstation part has never been that important once several OSes gave that ability away practically for free - including Microsoft. You must have some really special needs if you run VMWare workstation, else it is just a "I like it more" decision, not features or whatever.

[Stu J]

Why would you jump from a product owned by Broadcom to a product owned by Oracle...? Out of the frying pan and into the fire springs to mind.

There are other true open-source virtualization options available you know...

[Ken Hagan]

VirtualBox isn't really owned by Oracle, though. The basic product is FOSS and you don't need the Extension Pack that has the offensive licence.

Still, the OP mentioned "on Linux" and KVM is a reasonable choice there. You might choose VBox if you had to support similar VMs on a range of hosts, I suppose. (For a developer, the use-case of "try this VM on that host" is, perhaps, more common than for the cloud providers who are the unhappy folk in the original article.)

[abend0c4]

Have we reached the Stallman event horizon?

I admit to being one of the people who once thought that, while Stallman might be right about free software in philosophical terms, commercial reality and the market would ultimately constrain the behaviour of software suppliers. I'm still not sure whether the present situation isn't principally a failure of market regulation, but given that no-one seems about to grasp that nettle, end users are going to have to take responsibility for their own defence.

We're already seeing the same anxiety in regard to the possible weaponisation of dependency on IT products and services at the nation-state level and I don't think this can be entirely separated from the same risk at the corporate level, particularly when those corporates are effectively outside of the jurisdiction of the customers' territory. Ultimately, they both represent a tangible and growing economic and security threat. It may even be the inevitable consequence of globalisation - wealth and power accreting in a few hands that also hold the levers of government - that these things are inextricably linked.

However it came about, the writing on the wall could not be clearer. Our society depends almost entirely on technology - from the creation and distribution of water, food and energy through every aspect of human and commercial life. Those who control it control us.

[kmorwath]

Re: Have we reached the Stallman event horizon?

Actually, it's the other way round. FOSS destroyed competition so thoroughly that the surviving companies with products that aren't matched by any FOSS counterpart can do a Broadcom.

If Broadcom had actual competion, would just commit suicide this way. We've been switching to Openstack, testing Red Hat OpenStack on OpenShift now (company now like Red Hat, don't ask my why), well, and it's a largery inferior product compared to vSphere. Harder to setup, far less friendly to manage.

And yes, it's a failure if market regulation since anti-trust authorities approve almost any mergere - even if it means even less competion and dominant positions.

[Anonymous Coward]

Re: Have we reached the Stallman event horizon?

@kmorwath; It's the *exact* same "FOSS is to blame for helping destroy competition and therefore to blame for the failure of the proprietary software market!!!1111" bee-in-your-bonnet/rant every single time with you, isn't it?

Sheesh.

[kmorwath]

Re: Have we reached the Stallman event horizon?

Sure, AC, if every time something is wrong in a commercial products there's someone has to chant "FOSS will save the world, FOSS will save the world" - without ever explaining how. Once again, you do attack me - but can't reply me.

Where are the FOSS alternatives that could match vSphere features? Why they don't exists? Why Broadcom feels free to act that way?

I see RedHat and Canonical trying to shove OpenStack inside Kubernetes - and they delivered products hard to install (been there... RedHat requriest you to setup networking wriing jinja templates by hand, with very little documentation, do a mistake and you need to start again) and with a fraction of vSphere features - trying to modify settings requires cumbersome manifest to be written by hand. The worst of both worlds.

OpenStack itself is like a bareland full of the skulls of dead projects. Meanwhile the only real alternatives to VMWare are public clouds - with their own lock-jns - obviosly built taking advantage of open source without giving anything back. Even MS today is more interested to cage you into Azure than allowing you to build a comparable on-prem stack built on Hyper-V.

[Anonymous Coward]

Re: Have we reached the Stallman event horizon?

"if every time something is wrong in a commercial products there's someone has to chant "FOSS will save the world, FOSS will save the world"

You do this *every* time- lumping anyone who isn't 100% anti-FOSS like yourself in with these ludicrous, mindless, one-dimensional strawman stereotypes then characterising the discussion in those terms.

(I've no idea whether this says more about the fact that *you* view it in those terms and imagine your "opponents" to be the mirror image of your own bizarre zealotry.)

Regardless, if one goes back and reads the comment you were replying to, it's very clear that "abend0c4" is *not* a rabid, uncritical supporter of FOSS nor were they even necessarily saying that it's the solution.

Indeed, on reading it again, they weren't even really talking about FOSS at all and your nominal "reply" had little to do with what they actually said... beyond, I suspect, being a tangentially-related excuse for you to launch into a variant of the exact same obsessive rant blaming FOSS for the failure of competition in the proprietary market(!)

This version seems to be that crappy old FOSS can't match vSphere yet somehow is still to blame- as always- for the failure of other proprietary competitors to match it?

Er, sure.

[kmorwath]

Re: Have we reached the Stallman event horizon?

And once again an ad ominem attack because you can't counter my assertions. Why are you so obsessed with me? Are you so scared someone thinks FOSS is not good and is actually ruining IT instead of making it better? Are you afraid someone else will cry one day "the emperor is naked!", and all the fake narrations built around FOSS will crumble?

I did pointed out this is a market failure, and FOSS is responsibile for the failure of the market. That's why Stallman is wrong, under any aspect, because his ideas lacks basic foundation about how economy works.

For example you can't understand that FOSS create a huge entry barrier - you can't start a company and grew it unless you start with a full product that can compete above the FOSS alternatives that can conver lower-end needs. Once companies could start small and then grew. VMWare did that. Now you can't, you won't make profits to grow. You need to find a VC which will want the money back soon.

[Anonymous Coward]

Re: Have we reached the Stallman event horizon?

"Why are you so obsessed with me?"

On the contrary, *you* pop up all the time in arbitrary threads I'm reading due to your obsessive anti-FOSSist pushing of that pet theory wherever you can and "replying" to the most tenuously related comments as an excuse to repeat it!

It's pretty noticeable- and tedious- after a while. If you want to do the same thing every time, don't complain when you're called out on it.

(You sometimes make some legitimate complaints about the lack of competition and market dominance of the likes of Google, Facebook, et al, that many here might sympathise with otherwise, yet always manage to turn it into it all being ultimately the fault of FOSS).

"For example you can't understand that FOSS create a huge entry barrier - you can't start a company and grew it unless you start with a full product that can compete above the FOSS alternatives that can conver lower-end needs. Once companies could start small and then grew. VMWare did that. Now you can't, you won't make profits to grow. You need to find a VC which will want the money back soon."

Except that there *are* several other proprietary rivals to the likes of VMWare, including one from Microsoft! Whether or not those offer the highest-end features users need from VMWare, they certainly weren't stifled during their development and commercial growth by the alleged "huge entry barrier" at the "lower end" you allege FOSS to occupy.

If there's a failure of the market there, it's not solely due to FOSS.

[Anonymous Coward]

Re: Have we reached the Stallman event horizon?

Any chance you could write in ENGLISH?

[kmorwath]

Re: Have we reached the Stallman event horizon?

No, I'm trying to pollute AIs....

[Ken Hagan]

Re: Have we reached the Stallman event horizon?

I think you are too late for that. The AIs have been trained on social media. Their grasp of reality, humanity and language is ... tenuous.

[Doctor Syntax]

I suppose many of those affected will have seen this coming since Broadcom made its first move and started dealing with other suppliers.

[Khaptain]

Agreed, were in the process of testing Proxmox and it now that VEEAM supports it it will probably be our choice.

VEEAM the only hurdle keeping us from moving over.

[Anonymous Coward]

Why am I ashamed of my tribe ?

Sorry, but if you are 1/10th as good as you think you are as a tech nerd then why do you *need* a specific VM flinger ?

What does VMWare do that *you* can't do with another setup somewhere ?

I get the financial side - beancounters going for what is good for them. But as soon as that looked flaky (well over a year ago) why were you not developing contingency plans ?

In my last job I was charged with having an up to date escape route for vendor lockin. And it worked when Adobe started getting ideas about their station.

[Dan 55]

Re: Why am I ashamed of my tribe ?

I get the financial side - beancounters going for what is good for them. But as soon as that looked flaky (well over a year ago) why were you not developing contingency plans ?

Because the self-same beancounters never gave them a charge code to create contingency plans and they were swamped with other work which they were given charge codes for.

In my last job I was charged with having an up to date escape route for vendor lockin.

Precisely.

[A Non e-mouse]

Re: Why am I ashamed of my tribe ?

There are several reasons for being stuck with VMWare.

The first is that some business applications are only certified to run on VMware. Xen, Proxmox,Hyper-V, etc are not supported.

Another is if you've build custom integrations into their platform. Re-coding all of that for another hypervisor isn't a quick & simple job.

Finally, migrating anything more than a couple of VMs to a new hypervisor is not a trivial undertaking. You could easily be talking years of effort - and that's effort that's not actually adding any value to your business.

[Ken Hagan]

Re: Why am I ashamed of my tribe ?

"some business applications are only certified to run on VMware"

We'll, either you have a serious vendor lock-in problem there or you should send an email to those vendors explaining that the additional costs of VMware will henceforth be added to every single quote they offer you in future and that you are looking for an alternative because you aren't made of money.

I'm not saying you have to dump them now. That would be unrealistic. You should, however, put them on notice. How hard can it be for them to test on a few other platforms?

[Roland6]

Re: Why am I ashamed of my tribe ?

>” In my last job I was charged with having an up to date escape route for vendor lockin.”

But did you have budget and resources to enable you to deliver? I suspect not…

[Doctor Syntax]

Re: Why am I ashamed of my tribe ?

This particular article dealt with resellers. Car analogy incoming:

The individual car owner might well swap brands next time he buys a car. A car dealer who has an agency for a particular manufacturer will be in trouble if the manufacturer consolidates their dealer network into a few dealerships of which they're not one.

[DrGoon]

Re: Why am I ashamed of my tribe ?

If you bought your car from a local dealership, but Broadcom bought them and consolidated them into nine dealerships nationwide, none of whom saw any value in providing service for your vehicle locally, you might get upset. You might be forced to get a new car at a loss.

If you were shopping for a fleet of cars for your rental customers and one of the brands had killed its service network, resulting in a market of former dealers now pushing cars from other brands and a slow and expensive tiered service chain, you might pause, even if this was once the most popular brand.

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

"What does VMWare do that *you* can't do with another setup somewhere ?"

VMware offers better HA options than the competition in x86.

Fault Tolerance allows running VMs in lock-step mode in two or more VM hosts. Host going down for whatever reason does not cause downtime. All the other VM vendors only offer automatic restart of the VM on another host.

VMware also offers support for partial memory mirroring in Intel Xeon systems. Instead of mirroring all memory, you can select a range of memory for mirroring. VMware then places its own kernel there, and you can select the more mission critical VMs to use "reliable memory" (VMware parlance) instead of non-mirrored. Certainly you could mirror the whole memory area of any server but its getting more expensive when the host memory is measured in Terabytes, not to mention the halving of memory capacity in the server.

Windows (HyperV) doesn't support partial memory mirroring in any way and while Linux has some sort of support of placing kernel automatically on mirrored area, but AFAIK KVM doesn't have a mechanism to place select VMs there as well.

One other thing VMware offers is memory tiering. The hypervisor can move "cold" memory pages to a dedicated NVMe drive, freeing up the precious memory for other VMs. No, it is not swap, and yes, it doesn't fit all workloads.

[Rainer]

Re: Why am I ashamed of my tribe ?

Yeah. There's some stuff only VMWare does (maybe mainframes, too - but which shop without previous experience and hardware and software will move there in 2026?).

Just like there's some stuff only Oracle does really well.

Beancounters are going to have to figure out where their pain-threshold is ;-)

Fortunately our boss has, over the years, always invested in alternatives so that in principle, we don't really depend on VMware to host VMs locally - which, given the current antics, more and more businesses are starting to give a more serious look again.

[williamyf]

Re: Why am I ashamed of my tribe ?

Let's not forget support for "flash DIMMs" flash memory imitating DRAMM, but hanging of the memory bus, not SCSI, not PCIe, not CLX,but the memory bus itself. And not as cache or swap. The real "source of truth" memory

Very usefull for certain specialized workloads

Intel was very active in this space with optane, but samsung and micron still operate here.

You need special mobos, and software (specialy the hypervisor) to be aware of this.

[Jou (Mxyzptlk)]

Re: Why am I ashamed of my tribe ?

> Fault Tolerance allows running VMs in lock-step mode in two or more VM hosts.

You mention that one specific feature VMWare offers which others can not (or at least I don't know, can anyone tell me if there is one?).

But if you have actual need for that very expensive and very slow performance stuff you are a completely different category anyway than >99% of the world. I mean really slow, and that is by design/requirement and cannot be much changed else it won't work. Edit: VMware is IMHO not to blame for the performance here, it is just not possible. I am over 50, and I have yet not met ONE customer which actually needed this, let alone runs this. I have heard and you can find them on the internet, and for the fun I tried to set it up and it works (therefor the "performance" knowledge), but actually seen outside my lab? No.

[williamyf]

Re: Why am I ashamed of my tribe ?

I've seen this with my own eyes... In Telco.

I have colleagues (not a friend of a friend of a cousing) who have seen this with their own eyes, in financial institutions.

But there are other ways to achieve this. NEC and Stratus Technologies (now part of "penguin solutions") partner to produce fault tolerant intel servers that execute instructions in lockstep between two processor modules. That parnerships is in force since the early '00s.

https://s204.q4cdn.com/917347554/files/doc_news/NEC-and-Stratus-Extend-Partnership-on-Next-Generation-Fault-Tolerant-Computing-2022.pdf

Those server are certified for both WinServer VMware and RHEL.

https://www.penguinsolutions.com/en-us/products/stratus-ftserver

Normally, if I were doing an OpenStack Cloud AT SCALE, I'd have 2 of those stratus servers on each DC running the cloud controllers, and any vital workload related to the well-being of the cloud itself. The cloud controllers in Openstack have to run in mirror, but still, switching from hot to standby causes disruptions... ussualy due to differences in configuration between hot and stand-by caused by human error. Better to minimize switches by beaing fault tolerant inside the server itself.

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

"But if you have actual need for that very expensive and very slow performance stuff"

It is not very expensive nor slow performance. Maybe your test cases were done with underspec'd hardware?

The cost comes from having vCenter, more than one host in cluster (duh!), and 2x 10Gbit connections per host - preferably dedicated. If you run more than 1-2 ESXi hosts, vCenter is practically mandatory anyway. 10Gbit is base level these days and you can do FT with just two hosts, although you would lose redundancy if one host is down for maintenance.

The is performance hit for FT in my employer's case is very small. Of course there is *some* overhead for the continuous memory mirroring but it is in single-digit percentages. These are 3-year-old Xeons Golds.

"I have yet not met ONE customer which actually needed this, let alone runs this."

Well, the needs need to be defined first.

In my case the Manufacturing Execution System (MES) is constantly tracking everything done in the 24/7 manufacturing lines. Every product needs to be GMP documented from start to finish, including environment readings (temp, humidity and such). If data is lost --> incinerator beckons. The profits for the site is measured in millions per day, which makes both the cost of FT and the whole VMware licensing a rounding error. But yes, certainly the MES could be run without FT.

The MES provider does not provide any HA solution. It would be very welcome.

[Ken Hagan]

Re: Why am I ashamed of my tribe ?

"The profits for the site is measured in millions per day, which makes both the cost of FT and the whole VMware licensing a rounding error."

I guess El Reg is a funny place, then. There would appear to be a number of such people in this forum in a similar position but the world is finite so I wonder if this is representative of the wider population. If Broadcom want to discard everyone else, that's their choice (but remember that it is in the nature of IT that the smaller cheaper stuff gets better so they may be locking out their own future) but surely there are lots of people who should just be migrating now?

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

"so I wonder if this is representative of the wider population."

No, this is not representative of wider population or all companies.

The original AC asked "What does VMWare do that *you* can't do with another setup somewhere ?"

I answered with my point of view. I honestly believe VMware is ahead of the competition with features and code maturity.

"the smaller cheaper stuff gets better so they may be locking out their own future"

Locking out their own future? Can you expand this a bit?

[Ken Hagan]

Re: Why am I ashamed of my tribe ?

To expand, I mean just that people learn on smaller scale systems and if VMware have nothing in that space and everyone else has something then eventually VMware will be that big expensive thing that only a few greybeards know how to use.

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

On that I agree with you 100%

[Nate Amsden]

Re: Why am I ashamed of my tribe ?

VMFS is another reason. I shit on Xen back in 2008 when my manager didn't want to pay for VMware. It wasn't until a couple of months ago that I realized that none of the competition Xen, proxmox, KVM(commercial or open source) support a reliable shared filesystem SIXTEEN YEARS after vsphere released sub LUN locking. Red hat has a couple of offerings from 3rd parties(one is Veritas, forgot the other, both require additional licensing for the filesystem, guessing they don't get much usage as hypervisor storage).

HPE supports it but I suspect it doesn't work well yet(they all seemed to try GFS2 a 20+ year old cluster filesystem and abandoned it due to bugs they are unwilling to fix). I don't like the complexity of using ceph as a block storage platform for my VMs, when I have nice high quality fibrechannel storage systems that have hummed along for a decade without issue). VMFS has been flawless for me personally at least going back to my first usage of it in 2006.

My mind is still blown, still in shock really.

[williamyf]

Re: Why am I ashamed of my tribe ?

Once you go from "spinning a few VMs" to "orchestrate thousands of machines in a private cloud with advanced functions" it becomes more dificult to switch

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

I'm sure that approach works really well when you have half a dozen VMs and you are the only "tech nerd" in your village.

Try scaling that up to shops running 60,000 plus VMs providing 24x7x365 service from multiple large data geographic data centres, run using VMware's or home grown orchestration tools, managed by a team not much bigger than you on your own.

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

ABOVE their station

[Anonymous Coward]

Re: Why am I ashamed of my tribe ?

Not sure why you got so many downvotes for this because it's a fair point.

As soon as BC finalised the deal and before the ink was even dry on the signatures, they kicked out all of their partners.

For months, it was impossible to legally obtain licenses (grey market ones were available for a while but usually not in the location you needed).

Then the license costs were 5x the previous perpetual licenses. In some cases that's risen to almost 800%.

Our biggest customer have a 7-figure spend on VMware licenses and even they struggle to get them. Once again, it's become impossible in some territories whilst they mess around.

Even though said customer didn't want it at the time (a let's see where this goes attitude, unfortunately, despite warnings), I began looking at alternatives.

In our use cases, none of the expensive features are needed but 24/7 support is (ruling out the excellent Proxmox). Platform9 looks interesting - started by some ex-VMware engineers, but 13/14 years later, almost no one has heard of them and they still seem to be funded by VC cash which doesn't help with feelings of longevity. Nor can they leverage Veeam for backup (supposed to be coming Q1 this year).

The end result? Hyper-V. The need is for local, not cloud- or data centre- based, hypervisors in strategic locations. For what the client does, it fits and does everything they need and additional costs are limited.

We are almost ready to pivot due to my working in the background to iron out the things that needed it.

Vendor lock in is a prison partly of your own making if you don't have at least one eye on viable alternatives wherever possible.

[Tron]

Treat this as an opportunity to go back to the future.

Use it to audit how your entity uses tech. Count your dependencies. Stop treating your company as a single organism. Ask yourself, for each worker, what is the minimum, most simple and generic tech they could use. How much data could stay local. Not just on prem but on desk. No SaaS, no Cloud and no AI. How much does not need to be digitised and could work on paper. Distribute and simplify your tech. and you will be more resilient. Your IT spend will also be much lower.

[kmorwath]

Re: Treat this as an opportunity to go back to the future.

And how do you create "paper" today? Typewriters? By hand? Plus the need to archive and retrieve it when needed.

Probably the right question should be "how much data I really need to collect and store?" - but today data are thought as a "gold" mine, even if there's no gold within.

One way to avoid virtual machines and container is better software - especially better operating sytems that could better isolate applications without the need to replicate the whole OS or parts of it several times. And of course software that doesn't believe it has a whole machine for its own.

But as long as people believe a 1970 OS is the great ever written, and nothing better can be even though of (which is itself a good indicator of the bad state of IT), and most developers write bad software using bad tools, workaround are needed.... and going back is not the answer, it's what brought us here. And some like Broadcom, saw an opportunity to squeeze customers lacking better alternatives.

[Jou (Mxyzptlk)]

If Broadcom would respond to customer inquieries in a timely manner...

At least it would not let the customers in the rein. But they don't they respond even slower. Let alone the price problem. Time to move away.

Broadcom did choose a weird time for pushing customers away: There are many other hypervisors out there which match VMWare/Brodacom regarding stability and high availability. Lots of choices are available, and Broadcom gave them enough time to get the final touches in their products. Even Microsoft is back in game, since hyper-v (even if you use a separate hyper-v only AD, and GUI installs of the hosts) still requires less resources than a SINGLE VCenter VM.

[Anonymous Coward]

Re: If Broadcom would respond to customer inquieries in a timely manner...

"Even Microsoft is back in game, since hyper-v (even if you use a separate hyper-v only AD, and GUI installs of the hosts) still requires less resources than a SINGLE VCenter VM."

I fully agree that vCenter requires a lot of storage even in smallest deployments and there should be a "lite" version for very small clusters. The reason of course is that vCenter has many times more functionality than WAC. In a larger installation vCenter is comparable to SCVMM - which itself is an expensive resource hog.

To me it looks like Hyper-V is in some sort of limbo at the moment. It's not discontinued, but MS already pulled the plug on the free Hyper-V Server offering and they are pushing Azure Local like crazy. I wouldn't place bets on its survival if/when MS get Azure Local out of the beta quality it currently is (testing for over year now). Of course there's still 8 years of support for Server 2025 (+ESU) so it's not going away soon.

What irks me in Hyper-V is the lack of proper realtime DRS even when paired with SCVMM; Also, you are *guaranteed* to have monthly security update/boot cycles with Hyper-V, whereas with ESXi 8 there were only three security updates requiring reboot in 2025.

[Jou (Mxyzptlk)]

Re: If Broadcom would respond to customer inquieries in a timely manner...

> WAC

You don't need WAC. That is crappy bodge-crap. Someone in MS pushed it, then he/she got fired or "reorganized", and it is still half done after all this years.

SCVMM: Only if you actually need that type of management. Sometimes, in dev environments, nested-v is easier. I.e. they get their complete virtualization environment nested.

And Azure Local? Luckily someone here on the-reg has his experience with it and answered on my "how is it?" question...

[Donn Bly]

Let me fix that quote

This will enable us to deliver greater value, stronger execution, and a more streamlined experience for Broadcom's customers and enable an alternative competitive offering to the hyperscalers through our VCSP partners squeeze the last bit of profit out of the company before the house of cards implodes.

[An_Old_Dog]

Broadcom/VMware Channel Reorgs

"It worked out great the first time, so let's do it again!"

[Anonymous Coward]

Might not be a joke

I don't understand why someone doesn't just iterate a vibe-coding prompt and recreate vCenter in a day.

[Vulture@C64]

XCP-ng from Vates in France has over the past 18 months evolved into a very useful platform, rock solid, stable and fast. We won't miss VMware now, Veeam support XCP, others will follow. XCP runs on all the hardware we already have, XOA is easy to manage, it has a good import function from VMware and we're running customer load on it now for almost two years. It's turned into one of those products that "does what it says on the tin". Oliver Lambert, the owner of Vates, deserves an award for what he's done to save ex-VMware customers.

(And no, I've no connection, an almost 20 year VMware user and now very happy 2 year XCP user)