Broadcom’s VMware strategy looks ever more shaky - and less relevant For many, VMware by Broadcom has meant misery by the boatload. The virtualization platform's new owners have embarked on price hikes for the big and forcible eviction for the little. The dividing line isn't clear. A 24,000 VM migration by share repository Computershare seemingly triggered by the gouge suggests things might not …
The only strategy that I can see is either pay up or piss off.
This time though it looks as if though will turn a large business into a niche market.. New up and coming companies won't be able to afford the new prices so they will look elsewhere..
Alternatives are either Nutanix or HyperV... Anyone with Windows inhouse is going to seriously start to look at Hyper-V.
Other solutions exist for smaller (and larger) installations, such as Proxmox. For tech businesses, there's the DIY approach to open stack*. In the carrier space, companies like Huawei and Ericsson provide off-the-shelf private cloud solutions (though I would be inclined to avoid both of those).
* The learning curve is potentially reduced because you can ask a chatbot.
While it might annoy customers, saying that it's not a viable strategy ignores the history of IT - all the big companies have been doing this for years.
Or have I missed where HP, Cisco, IBM, Dell, SAP, Oracle etc acquired other companies and improved the products rather than just ramped up prices while reducing operational costs by reducing features?
And yes, I'm old and grumpy and my glass might have been half full if it wasn't for evaporation...
If you move a virtual machine from VMWare to HyperV, then what else do you have to change?
Anything accessing the services on that VM isn't going to notice the difference.
Software running on the VM most likely won't notice the difference.
The only thing that will really notice the difference is your VM administration tools, clustering, backup, and stuff like that. Compared to say migrating from Oracle to Postgres, that is pretty painless.
If you have GPU workloads, then HyperV maybe isn't the right solution for now. I have one GPU workload on HyperV, and it takes an entire graphics card via PCIe passthrough. Server 2025 looks like it will be an improvement in that department, but obviously you aren't going to put that in production right now.
Broadcom only saw the balance sheet and didn't understand the business. At the end of the day, their products are centered on running applications on x86 operating systems, and that's nothing special. Outside of the first few years when they were pretty much the only game in town for x86 virtualization, vmware's products have never been their competitive advantage. It was everything around it... from their own sales and support teams, brand recognition, the ease of purchasing their products, consultants, third-party backup tools, service providers that didn't mind paying the vmware tax, storage vendors that knew they could sell a hundred thousand units because their products were vmware-certified because hundreds of thousands of businesses ran vmware. Broadcom killed all that day one. The world that was built around vmware can't be sustained on those 500 customers Broadcom wanted to plunder and pillage.
The winner in all of this will be determined by the companies that make backup software.
Gouging has become the preferred pricing strategy in the US because the MBAlers love it. It's supposed to be indicative of a functioning market. But, as we see here, the market is not really able to respond with more supply so customers have no option but to pay up. I don't think Broadcom is unusual in this, except about their honesty.
I expect to see more of this as companies move more of the IT processing to the big 4 (or 6 if we think about Oracle and SAP). We're seeing this already with Microsoft's decision to move everyone onto subscriptions. At some point, once they think they've reached market saturation and sufficient lock-in the cloud providers will start raising prices in lockstep. We should set up a sweepstake for the excuses they give: AI; electricity costs; a new island for the boss, etc…
... summing up nicely the situation, aka, broadcom are locked into the huge debt and the obligation to rip customers off.
They're screwed in my opinion...
"An equation of value replaced by an equation of pain."
Sounds like a nice description of my marriage...
Yes, I'm now happily divorced.
Yes, it was very tough as all heck but I never regretted it ...
Is not Farage the main board member's advisor of Broadcom by any chance? He managed to fukc whole UK up with BoJo with the 350M lie. This great ideas of Broadcom to rise prices just as bad as brexit has been. Everyone is moving away from it just like happened to Novell in the 90's. They close licensing, no one ever managed to clone, hack etc and they ended up sitting on SuSe and some old Unix licenses that nobody cares or mentiones anymore. 90% of the coders nowdays don't even know what Novell is or was. The whole bloody Active Directory was stolen from Novell by M$crosoft!
So, this will be VMware and Broadcom in 20 years just like Novell. Although Novell was great and didnt fukc up their customers, but it was way too closed to properly spread.
Anyhow, Broadcom could have make it bigger than ever but nope. Broadcom become again a killer, Symantec killer and now VMware killer too. Two of my well respected and loved companies that made my carrier path very well. Amen to them!
I've been playing around with proxmox and for the small users that I support ( that only need a Domaincontroller/dns/dhcp server and maybe a file and print server) this could be ideal. Of course if using Microsoft it may be nbetter to use HyperV
Even moving them to Microsoft 365, and using Azure connect is better than relying on VMware now.
It's a nice compromise if you only need a few VM's in a client environment and you can build a supported host. Vendors like Supermicro https://www.supermicro.com/en/solutions/azure-stack-hci are very reasonable for this. You can even run Azure Virtual Desktop from it, as long as you have the user M365 licensing and appropriate networking.
Windows licensing is tied to an Azure subscription https://azure.microsoft.com/en-us/pricing/details/azure-stack/hci/ on a physical core basis.
Ive used VMWare for Windows virtualization in my Mac for years.. when the M1 was released, VMWare compatibility was zero. I waited for months after moving my workload to an M1, even made an 'upgrade' purchase thinking the next update will include apple silicon compatibility. In the end I think it took a couple years after the release of M1 before Vmware was able to release its compatible software. Too late, as I've already ditched them in favor of Parallels (who made it a point to be compatible with apple hw on day 1).
well...I don't think VMware or Broadcom cares a lot about the desktop user running Windows on their Mac. That wasn't their bread and butter... They literally gave away server software much more powerful (their ESXi product --may it rest in peace). They really didn't care about that. If someone ran VMware workstation or whatever they called it on their Mac or PC and that influenced them go go with VMware in their data center, great, however unlikely it might be (though I'm sure a few people who ran VMware Workstation assumed they were now VMware experts). If they snag a few extra bucks from their big money clients for little stuff for the desktop, fine. But it was all lost in rounding errors on their bottom line.
But I agree with your overall statement -- VMware pre-Broadcom was abusing their customer base with license rule changes, cranky fat clients and when they finally came up with a web client, they made probably the very last big Adobe flash application when all the rest of the world had realized Flash was a Bad Idea and HTML5 was good. And, I've never forgiven them for allowing a time bomb intended to keep testes from running beta software "too long" to slip into a release (Aug 2008?). They owned the market because no one got fired for buying VMware...and they knew it and it showed for well over a decade.
Broadcom just makes it so painful, now people are going to get fired if they don't get off VMware. That's really a stunning accomplishment, but as the article indicated, all of VMware is now a rounding error in Broadcom's business.
We're using a couple of Broadcom's enterprise products such as VMWare and Automation, we're working on our exit over the next 18 months 'cos despite cloud vendors being expensive it still works out 75% cheaper to migrate, including the pain, time and effort to do it.
For enterprise customers already in the RHEL ecosystem, they've extended their K8S offering to include the KubeVirt bundle to allow running full VMs there. For larger shops already working in containers, this may offer an off-ramp from Virtzilla's impending price gouges.
yeah, there are options, but ... if I were in the RHEL ecosystem, I'd be looking at their history of support before IBM ("you need help with your license problems, we can do that...anything else? Better off with google"). their current owners (IBM) and their reputation for careful product maintenance (that was sarcasm), confusing "Adding Value" with "Raising Prices", and their recent jerking around the CentOS community, RH would NOT be the direction I'd be looking for long-term commitments.
If it wasn't for Broadcom stealing the show for customer abuse, we'd probably be talking about IBM/RH abusing their customers.
Broadcom has a 22 different divisions, many of which are not as profitable or attractive as its chips arm. They will need to invest more in software "to grow" rather than milk old tech as they have been doing with VMware. The speculation of their acquisition of Zscaler is interesting (https://csec.substack.com/p/broadcom-to-acquire-zscaler?r=3xqwa9). They need to do more of this.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
