Euro-Cloud Anexia Moves 12,000 VMs Off VMware to Homebrew KVM Platform (theregister.com) 53
The Register's Simon Sharwood reports: Broadcom has lost another sizable customer for its VMware platform: Austrian cloud provider Anexia has moved 12,000 VMs, some of them rented by major European businesses, to an open-source system based on the KVM hypervisor. Anexia was founded in 2006, is based in Austria, and provides cloud services from over 100 locations around the world by placing equipment in third party datacenters. Clients include remote access and control vendor TeamViewer, and airline Lufthansa -- plus plenty more outfits that need reliable hosting and service to match.
CEO Alexander Windbichler told The Register that after Broadcom acquired VMware, increased licensing costs, and made big changes to its partner program, Anexia remained eligible to operate a VMware-powered cloud. But Windbichler felt he couldn't afford to continue, because Broadcom offered new terms that saw the cost of VMware licenses rise sharply. The CEO preferred not to enumerate the increase precisely however The Register understands it exceeded 500 percent. Whatever the actual figure, Windbichler said the cost increase "Would have been existential for us."
"We used to pay for VMware software one month in arrears," he said. "With Broadcom we had to pay a year in advance with a two-year contract." That arrangement, the CEO said, would have created extreme stress on company cashflow. "We would not be able to compete with the market," he said. "We had customers on contracts, and they would not pay for a price increase." Windbichler considered legal action, but felt the fight would have been slow and expensive. Anexia therefore resolved to migrate, a choice made easier by its ownership of another hosting business called Netcup that ran on a KVM-based platform.
CEO Alexander Windbichler told The Register that after Broadcom acquired VMware, increased licensing costs, and made big changes to its partner program, Anexia remained eligible to operate a VMware-powered cloud. But Windbichler felt he couldn't afford to continue, because Broadcom offered new terms that saw the cost of VMware licenses rise sharply. The CEO preferred not to enumerate the increase precisely however The Register understands it exceeded 500 percent. Whatever the actual figure, Windbichler said the cost increase "Would have been existential for us."
"We used to pay for VMware software one month in arrears," he said. "With Broadcom we had to pay a year in advance with a two-year contract." That arrangement, the CEO said, would have created extreme stress on company cashflow. "We would not be able to compete with the market," he said. "We had customers on contracts, and they would not pay for a price increase." Windbichler considered legal action, but felt the fight would have been slow and expensive. Anexia therefore resolved to migrate, a choice made easier by its ownership of another hosting business called Netcup that ran on a KVM-based platform.
Broadcom doesn't care (Score:5, Insightful)
Broadcom doesn't care. The revenue they lose from companies leaving will be more than offset than the revenue they squeeze out of companies that can't so easily leave.
Broadcom knows that VMware doesn't have much of a moat anymore so Hock has decided it is time to kill the Golden Goose.
It's brutal but it will likely maximize Broadcom's return on their purchase of VMware.
Clearly Hock buys into the maxim that the only thing that matters is increasing shareholder value. No other stakeholders matter.
Re:Broadcom doesn't care (Score:4, Insightful)
Of course, longer term the prospects for that sort of customer base are limited.
Sure, IBM made a big business out of it with mainframe, but moving off of that actually requires porting software. Meanwhile the VMWare competitors are able to run the exact same application software.
Sure there's migration work and companies are reluctant to bother, but the "stickiness" is far less sticky than IBM enjoys with mainframe. Even less sticky than proprietary Unix systems, and those largely failed to retain their customer base. So if vmware becomes some weird niche of only a few very reluctant companies, it's going to lose it's appeal pretty quickly.
Re: (Score:3)
It likely can be pretty easy to move off of VMWare. One can use a tool like Commvault which backs up VMWare VMs, but can restore them to another platform [1]. Other tools can import VMs via their disk files, which Proxmox can do.
The hard part would be cloud stuff, due to serverless services, and how VPCs are arranged, but even then, that is relatively easy compared to getting stuff off a mainframe architecture that has been there for decades.
[1]: Wish Red Hat stuck with RHV/RHEV. OVirt could easily have
Re:Broadcom doesn't care (Score:4, Interesting)
With you on the ovirt thing. RH got distracted by openstack and the promise of having on-premise follow cloud-like models for virtualization, which largely didn't happen (turned out that the general concept was overly convoluted for internal IT, and even to the extent it might have been desired, openstack was kind of always half baked at best). They gave up before really even trying to compete directly with vmware, which could have been interesting. Now they've given up on openstack and are trying to make openshift a thing including virtualization, despite not really being a natural fit for that use case either, rather than admit they've really just given up on being a serious virtualization host period.
Re:Broadcom doesn't care (Score:4, Interesting)
Nail, head hit. If the world revolved around Kubernetes, Red Hat would be doing things right. However, most VM farms are generally "pet" VMs, as opposed to "cattle" to be used for an app server, chucked and another one cloned with a TF config stamped on its forehead and moved in its place. Kubernetes tends to be handled by other means these days... often in the cloud.
What oVirt/RHEV brought to the table was a virtualization system that even (IIRC, was told to me at a trade show) Microsoft supported as a platform for Windows. KVM is a solid hypervisor, and RHEV brought a good management plane. Had they pushed that as an offering with an easy way to migrate from VMWare, Red Hat likely would be part of a Big Three when it comes to enterprise virtualization.
OpenStack was dead in the water. Off and on, I was asked to do OpenStack projects and compare it against other virtualization solutions. Yes, it would work if you tossed a ton of cheap manpower at it, but other than IBM, I have not seen any OpenStack stuff after IBM bought SoftLayer and "nudged" RH that direction. OpenStack has some issues with its architecture. For example, if Keystone has any issues, or crashes, say good-bye to your overcloud... and undercloud.
People didn't want that complexity. VMWare and Hyper-V are relatively simple. Get backend storage going, be it NFS, SMB/CIFS, or iSCSI, toss the OS on a number of hosts, admin from a console. Done. The interactions between machines are relatively simple, and are robust. This is where Proxmox excels at, since getting machines installed with it is quite easy, although I've not really messed with Ceph as block storage, and probably would use CIFS or NFS.
Re: (Score:1)
Sure there's migration work and companies are reluctant to bother, but the "stickiness" is far less sticky than IBM enjoys with mainframe.
That's because when those on mainframes look at the alternatives, they start to see that too many of those alternatives regard 3 years as long term. And they go "fuck that" because for them 30 years is long term. And 3 years could be how long it would take to just do proper research on what it might take to migrate all their zillions of lines of code on their thousands of internal apps to non-mainframe platforms.
Too many frameworks and platforms nowadays have a typical lifespan of 18 months before some back
Re:Broadcom doesn't care (Score:5, Insightful)
I am not impressed by Broadcom's strategies. They inherited a ton of IP with Symantec, stuff that is extremely useful and relevant. For example, if they got businesses using Symantec Encryption Desktop, it would be a very well-received DLP that is easy to use, secure, but yet recoverable (with ADKs) in case of an employee leaving.
Broadcom should be seeding VMWare everywhere. Instead of killing VMUG Advantage, they should just get rid of serial numbers and allow unlimited downloading so people can happily run vSAN and other stuff in their homelabs, and then scaling license fees to the uses, from a free edition for home/SOHO people to support for the top enterprises. Maybe take a look at Red Hat and how they are doing. Get VMWare everywhere and anywhere. Software doesn't care if it is with 100 installs or 100 million installs, so seed it, and eventually that will be made up by volume. The economy sucks, so if VMWare/Broadcom started adding cool new features, it would get people buying them.
Instead, it seems their model is the opposite. It works for the short term, but this is now making things like Proxmox and XCP-NG mainstream, with mainstream app support from Veeam. Proxmox is starting to gain enterprise capabilities, although in an alpha stage right now, and it is only a matter of time before they are on par with VMWare for most things.
As for Anexia, I'm curious how their KVM control plane is coming along. If it is good enough, they may be able to sell it and have a true killer app on their hand, if it can work as well as VMWare for large installations.
Re: (Score:3)
By this theory it seems like Oracle should have been milked to death about 20 years go. Maximizing the number of adoptions of Oracle certainly isn't their goal. Yet so far it has proven sustainable. At least, sustainable enough to outlast Larry Ellison.
Re: (Score:3)
Broadcom doesn't care. The revenue they lose from companies leaving will be more than offset than the revenue they squeeze out of companies that can't so easily leave.
Broadcom shouldn’t be so damn ignorant. It’s stories just like this that prove why they shouldn’t. If moving 12,000 VMs was found to be “easily” done, then the decision to leave Broadcoms greed will be easier than Broadcom is counting on.
Big mistake.
PSA slashdot using very malware-like ad platform (Score:5, Interesting)
It appears slashdot is now using an ad platform that is extremely intrusive and acts in a malware-like fashion with javascript that watches the page and reloads it if anything is removed or blocked.
If you try to block it with ublock origin, the site will completely break itself. I've been contributing to slashdot for many years (a couple of decades), and I feel like this move is a real slap in the face and I'm not likely to continue to come here and contribute if this continues.
urls involved in this ad scheme include html-load.com, css-load.com, content-loader.com, error-report.com, and 07c225f3.online.
Re: (Score:3)
It's ok with NoScript.
Re: (Score:3)
NoScript (on desktop Firefox) enables selective blocking of the domains. With adequate choices, it is possible to configure it to have slashdot behave properly (no ads visible, all javascript features usable as normal). I don't want to say too much, it was not difficult and the slashdot readership is clever enough to figure it out. Also I don't want to be banned. Personally I find NoScript to be very annoying to use so installing it was a last resort for me.
Re: (Score:2)
Also I don't want to be banned.
It's a sad state of affairs to be worried like this on mentioning something that should be so inconsequential.
Re: (Score:2)
It's ok with ...
NoScript, by default, blocks all JavaScript, even from the host server: Which of the 18 JS files are you allowing through?
I've found a little-used JavaScript manager (available for Firefox and Chrome), allows most Slashdot pages to load.
Here's an idea for Firefox: Container profiles that allow different extensions to load. For example, the default extension is NoScript but the "Shopping" container, instead uses the little-used extension.
Re: (Score:2)
Re:PSA slashdot using very malware-like ad platfor (Score:4, Informative)
I'm using ublock origin with no issues.
Re: (Score:3)
They aren't serving these unblockable ads to everyone it seems. But you'll get them sooner or later. From what I read, this ad platform targets specifically people running a blocker like ublock origin. It appears that injecting a script into the web page will stop the ads in their tracks. But I am unable to determine how to install this little script in ublock origin.
Re: PSA slashdot using very malware-like ad platfo (Score:2)
If you need to inject a script you use greasemonkey
Re: (Score:3)
I had an issue first thing this morning, but then I updated the uBlock rules and restarted the browser and it went away.
If it hadn't, then the solution would have been very quick and easy: to simply stop visiting Slashdot.
Re: (Score:2)
I've managed to block the actual ad content, but can't prevent the containers and resulting white space. If/when the site becomes unreadable, I'll bail completely, sadly, after many many years here. Shame. I imagine I'll somehow manage to go on living though...
Re:PSA slashdot using very malware-like ad platfor (Score:5, Insightful)
If/when the site becomes unreadable, I'll bail completely, sadly, after many many years here. Shame. I imagine I'll somehow manage to go on living though...
Same here. But fear not! I'm sure Slashdot's tens of daily users will dearly miss us.
Re:PSA slashdot using very malware-like ad platfor (Score:5, Insightful)
Disabling of new user registration means the ship is sinking.
Re: (Score:2)
Slashdot's decline began long ago when they started pushing Bitcoin [slashdot.org].
Any worthwhile alternatives if the /. ship sinks? (Score:2)
Are there any subreddits at Reddit.com with a feel similar to Slashdot? Not about specific tech areas, but a feed with a balance of both general and specific topics of interest to technical people and geeks, with a large enough following so that all topics get some discussion.
I've always wanted to enter Reddit as a regular lurker and small-time participant, but haven't found any feed with wide enough topics, and following a group of subreddits for all my interests feels impractical.
And I know of Hacker News
Re: (Score:2)
I'd say go to the nich-specific sites - for example for Mac news, 9to5MAC; for Linux stuff LWN or LinuxJournal; etc.
I'd agree that while HN good for finding some things occasionally, it's far too startup-focused, not to mention its overabundance of tech bros with overinflated egos about their 'opinions', including that of the site maintainers. Also HN is one of those sites that is 'too light' - there's simply not enough info on many of the 'stories' on the front pages (some are useless single-word titles).
Forgot to add... (Score:2)
Forgot to add the reason for going to niche-specific sites vs. somewhere that gathers all news: I've found that "jack of all trades" sites tend to die off eventually (Anandtech comes to mind), though it's also going to be true of any niche sites, I tend to find their discussions are more focused, and generally remain more active. As long as they don't start focusing only on the latest buzzwords [slashdot.org].
Re: (Score:2)
I haven't found anything else, yet, but my search intensifies every time /. turns this crap on (probably more development)...
Looks like a few people are working on workarounds, for /. and other sites, using userscripts in (violent|tamper|grease)monkey, but I haven't tried them, yet.
tinyShield [github.com]
Working around a broken ad-block block [aweirdimagination.net]
(Or google combinations of "slashdot html-load javascript monkey" ...)
Re: (Score:1)
Re: (Score:2)
Using noscript and allowing "nothing" gives me a big blank space at the top of the screen and the headlines are missing on the front page. You can still click on the space where the headline should be to go to the subpage (like this one) and the descriptions on the front page are still there. It looks a bit weird that way but it works.
Re: (Score:3)
Re: (Score:2)
Is this only for chrome's manifest v3 castrated version of ublock origin, or does it include the fully functional variant?
Re: PSA slashdot using very malware-like ad platfo (Score:2)
It doesn't work great with Firefox either. I had to mess around with ublock and noscript to get a working result.
Re: (Score:2)
Interesting. Maybe it's because I have excellent karma and the check mark to disable advertising always on.
Re: (Score:2)
Nope. I also have excellent karma and turned on the box.
Re: (Score:2)
Then it must be a regional thing, or they're doing A/B testing rolling it out only to some users.
Re: (Score:2)
Yeah, that makes sense. I started complaining about this before other people, so no surprise if they chose me as part of the first group since they have a love-hate relationship with me due to my volume but willingness to call them what they are.
Just a little while ago I tried to load the mobile site on my phone and it it didn't work. I was getting a little text about something like "hello this is a text bubble" and nothing else, with the page reloading over and over. I can still use it on desktop, though.
I
Re:PSA slashdot using very malware-like ad platfor (Score:4, Informative)
No this is on Firefox with ublock origin manifest v2. This is some very obfuscated javascript that Slashdot admins have added to their site that then pulls the ads in. They are using a service called ad-shield which markets itself as an "adblock recovery" platform. If they become more widespread this will definitely make large swaths of the internet unusable for many. It's a bit ironic because if the 30% of their viewers are using ad blocking, driving those 30% away won't in any way improve the revenue situation.
Re: (Score:2)
Where is the "Disable Ads" checkbox we had for 25 years?!?
Why would anyone be surprised? (Score:5, Informative)
Personally, I never understood why it made sense to pay tens or hundreds of thousands a year for functionality that you can effectively get for free.
Over the years I have had this sort of conversation with DevOps teams who were horrified by my attitude over this. They had to had to had to have their VMWare consoles for whatever reasons, and wouldn't even look at what was available open source. Much less the scripts I could show them that would do what they did in a more controllable way. Think of the training costs! And who are you going to call if something goes wrong! What if what if what if!
In some areas there is no winning over that mindset. So their employer ends up paying for their ignorance and just figures the amount into the P&L as software expense that everybody pays. Until they cannot anymore.
They come around only when forced to, as seen here. And even then the attitude won't exactly be "we will do more with less." They will do less with less.
Re: (Score:3)
Personally, I never understood why it made sense to pay tens or hundreds of thousands a year for functionality that you can effectively get for free.
This is why some companies do it. My most recent job was over a decade I spent with a company in the bottom half of the Fortune 500. A severance agreement with a bit still to go prevents me from talking about them by name. We used to use commercial Linux that we paid for. Why? So if anything ever went wrong with Linux, we would have some other company to blame for it. We had a lot of highly restrictive uptime agreements with customers with specific limited duration windows to do software upgrade
Re:Why would anyone be surprised? (Score:4, Insightful)
Personally, I never understood why it made sense to pay tens or hundreds of thousands a year for functionality that you can effectively get for free.
You gotta understand, young grasshopper, that from ~2004 until ~2007, VMware was the only game in town for Intel virtualization for large companies (that were not cloud providers) if you wanted it to run on Big Iron, and were not doing a "Science Project".
From 2007 to around 2012, the only credible alternatives were XenServer or Hyper-V, and none were as mature or as polished as VMware as a stack (i.e. the tools beyond the virtualization propper). It was around 2012 that other alternatives (including OpenStack) came to the fore in full force (and those were still still not as mature or as convenient).
Big companies that invested heavily on VMware are not able to move on a whim (even if they wanted to), and that is precisely why Broadcomm can jack up the prices and do all this nasty shit. Broadcom is moving like a hulking drunk to find the right balance that allows them to extract the maximum blood from the existing base, without losing customers en-masse. Given the analogy I used, you would see I do not think they are doing a good job of it.
And who are you going to call if something goes wrong! What if what if what if!
Sometimes your hand is forced on this issue. Your (cyber)Insurance, or some certification bodies in heavily regulated industries, may require platforms with recognized companies providing support for your platforms, Virtualization included.
So, even if developers and management were amenable to use FOSS without support (i.e. self-support), you can not do it. And granted, most management will not entartain theidea.
Re: (Score:2)
You're right about the certifications, but I have long held there are more than a fig-leaf fiction than any real risk mitigation. Yet the check signers believe. I don't have to believe if I am just taking their money and skimming some off and turning over to someone else.
Clearly the customers in this story have passed that stage that whether they believe or not, they can't afford it.
Re: (Score:2)
most leadership at companies does not prioritize the cost, I've seen many companies choose the most expensive option for 2 reasons:
1. support - companies are happy to pay a premium if they get white glove support if something goes wrong
2. ranking in their quadrant... they are happy to pay premium if the solution is ranked top in class
what you and I see as a priority or a cost saving opportunity are rarely aligned with the motivations/priority of decision makers that need to answer questions if something goe
Re: (Score:2)
Personally, I never understood why it made sense to pay tens or hundreds of thousands a year for functionality that you can effectively get for free.
In a word? Support.
I take it you have not effectively prevented your own proverbial head-chopping with a support contract in place at 3AM when the server shits itself during the last week of the fiscal quarter? It can be worth its weight in gold at the right time.
When you do NOT have a 1-800-FUCKED support number you can call at 3AM, the only one the boss is going to call (and blame), is you. No. Waiting on a neckbeard 7 time zones away to put down the bong and answer your urgent system down message on
Going to predict... (Score:5, Informative)
I'm going to predict that this year VMWare announces/previews a KVM based hypervisor as a successor to current ESXi.
Largely based on:
https://lore.kernel.org/lkml/2... [kernel.org]
Commentary has focused around this getting rid of the 'vmmon' kernel driver in workstation, but I suspect they have broader ambitions.
They are owned by Broadcom, and likely under a lot of pressure to stop doing undifferientad stuff. I would not be surprised if they view the ESXi operating system as fundamentally a liability rather than a differentiator, with the management stack and the firmware and userspace content being the differentiators compared to other KVM based stacks.
They have a different sort of kernel, and while they may be just happy with small customer base paying lots of money, the hardware vendors they need to run on will not be as excited to bother with ESXi drivers for the smaller and smaller opportunity. They already were challenged as it was with their previous market share, this makes things even less interesting for hardware vendors that won't see that big fat margin.
They always struggle with things like firmware utilities and updates, compared to their competition where that's super easy. VMware's play was make you deal with the BMC instead, but that limited them to the systems that bothered to flesh out their BMC, while competition just did not have to be so constrained.
Even without all this, it requires they actively develop and maintain their own distinct OS kernel and a big chunk of userspace that they just don't care about.
However, they can just instead have a linux kernel and get a minimal userspace maintained by a rich community and maybe have to tweak the scheduler, maybe have a few drivers. 'ESXi' suddenly becomes less to develop and their customers are just as stuck as ever because they care about the management stack and running unmodified VMWare images, which would still require VMWare proprietary virtual firmware and device emulation.
Re: (Score:2)
If they did that, I'd like to see VMFS as a filesystem under Linux. I don't like praising stuff, and when I do, I'm often damning with faint praise out of sarcasm, but stuff like VMFS and the fault-tolerant VM system where it runs two VMs at the same time are definitely gems.
VMFS can do clustered filesystems, all without all the overhead and admin time needed. Gluster2 and Ceph try, but VMFS is just point and shoot, and does all that automagically. Add vSAN, and you have encryption, compression, checksum
Re: (Score:2)
Those things would certainly come to "Linux' but I'd wager any bits that can't fit in userspace would be still proprietary closed source blobs with some sort of open source wrapper to comply with GPL (same how IBM and nVidia load binary blobs into kernel space by way of open source wrappers).
Ionos moved from VMware to Virtuozzo KVM last year (Score:3)
Cloud provider Ionos (formerly known as 1and1) moved from VMware to Virtuozzo, a KVM-based platform, last year. I used them originally because I assumed that VMware had a performance advantage.
Now that my Ionos workloads are on KVM I don't see any performance disadvantage of using KVM over VMware.
Enshitification of VMs (Score:2)
Enshitification is everywhere. It's the new business model. Introduce very low priced plans that entice users, tell them the new low price will be the new norm henceforth due to economies of scale, improvements in technology, AI, yada yada. Once enough customers throw their cottage cheese out of the window and sign up for endless supply of cloud-based cheese, increase the prices N fold, place restrictions of various sort. We now have a fledging fleecing business.
We promise cloud based AI will be different t