VMware's in court again. Customer relationships rarely go this wrong | Hacker News Reader

VMware's in court again. Customer relationships rarely go this wrong

theregister.com

249 pointsrntn197 comments

I was running about 1000 machines on VMware in my previous career. It was always a love/hate relationship with them. We were able to achieve a lot of our goals using VMware and it was hard not to be ecstatic about the results. At the same time, they were always a nightmare to deal with, the software was buggy and support wasn't great.

I always dreaded renewal time because it was normal for them to use it as an opportunity to extort us. Microsoft was a breeze in comparison. It's funny because Microsoft always had such a horrible reputation. I don't know if I was just so abused by VMware or what, but Microsoft was just easy. We had an annual true-up date and we always knew where we stood with them. We reported our numbers and that was it. No surprises ever and there was never an issue if we didn't report any growth. VMware was always pulling some kind of shit and was absolutely determined to push us over budget every time.

If I never have to have two weeks of meetings around discussing how many vCAN points we will need for the next year ever again, I'll be a happy man.

Microsoft is doing well and you were a small customer for them.

VMware on the other hand is dying because doing things that way hasn't been the state of the art for a long time.

Doing things that way (virtualization rather than containerization) fell out of vogue specifically because of how bad vmware was to work with. CPU quotas were probably what pushed serious people away from the product instead of machine licenses. I was early in my career but working with vmware products was the bane of our existence because if we wanted to make any sort of configuration change or spin up a test machine or really do anything at all it had to run through accounting which was just an instant non starter. we all started fiddling with alternatives and docker swiftly became reliable at least for spinning up a new web server or testing the latest and greatest whatever. vmware did this to themselves.

Exactly. You end up having to burn licenses for stuff like that, although I will concede that VMware always gave us more licenses than we paid for and they always included extra products. It was weird though. There was no rhyme or reason to it. One time they gave us 1000 licenses for VMware Fusion even though we didn't have any Macs. Microsoft, on the other hand, let us use whatever we wanted. If it was still around when it came time to true-up, we paid for it.

The vast majority of containers run on VM not baremetal.

But not on vmware, just zen and/or kvm with various management front ends.

VMware has always been a PITA, even in the late 2000's, we pivoted and bought several thousand physical machines for a new datacenter after they started to play tricks just weeks before we were going to turn up the DC.

They have always aspired to be Oracle like, where customers are hostages. Most people I knew who weren't stuck in the "Enterprise" trap moved to kvm/zen ASAP especially after the Westmere dramatically reduced the vm_exit() latency allowing for databases etc...

That was over 15 years ago, and outside of a very small number of niche use cases, tehre was no real argument to run container hosts on Vmware outside of a (IMHO) mistaken risk appetite.

It is really the fruit that ate itself, as had IT departments had a more data based risk assessment process, we would probably be heavily hybrid-cloud now. But the same Enterprise gravy train that VMware grew under killed them.

Shifting blame at great expense in licensing and agility to an _Enterprise_ solution was their jam...now Broadcom owns them an it is even worse.

Not in the F100. They're all VMs, all of the time, all on vSphere. Nutanix was the next best solution, with Hyper-V as a distant third. Hence why Broadcom ate them.

From what I hear and have seen, pre-Broadcom head-on VMware takeouts didn't go much of anywhere. But kubernetes-based (Kubevirt) products do seem to be having a degree of success.

Yeah, vsphere has a mile-long list of enterprise checkbox features that the sales managers can overwhelm the CIO's with on the golf course.

Kubernetes might have success, but AFAICS Kubernetes also sort of involves a new way of architecting applications (cloud native applications, 12-factor apps, microservices, etc.; whatever the buzzword du jour is). The idea with vmware was always to virtualize all those zillions of more or less idling physical servers, and get some snazzy management GUI to handle them all etc. etc.

Rearchitecting for containers is indeed a lot more effort and, indeed, one of the reasons for VMware's success was that it provided more efficiency without (at least initially) much in the way of operational changes.

But kuvevirt with Kubernetes does much of the same, especially for companies that are--or know they will--move to containerized workloads anyway.

kubernetes is where vsphere was circa 2010 or so. wide adoption with loads of room for growth and refining.

the vmware implosion + kubevirt gaining maturity is what will hyper-accelerate this.

nonetheless, it's a bit nuts that a vsphere alternative doesn't really exist for kubernetes. openshift, harvester and foss projects like portainer are close but not really the same thing. vsphere made it stupid easy to orchestrate vms at any scale with any level of experience. your cat could deploy a simple two-node vDC, and you could theoretically click your way through doing very very complicated things with your cluster (though you probably should).

esxi was silly easy to install most of the time also. provisioning kubernetes is finally (as of maybe three years ago?) pretty easy to do, but day two ops are still very command line. great for nerds like you and me; not good for former VIadmins who were suddenly thrown into the k8siverse.

this is even worse on the networking front. anyone who's messed with cilium or calico vs administrating NSX will know what i'm talking about. you can click your way through setting up a whole logical enterprise-grade networking fabric with vSANs, BGP, stateful firewall, the whole nine, all with near-linespeed performance throughout your vDC. cilium and calico, on the other hand, are powerful but have (to my knowledge) no real GUI equivalents.

storage is even worse. there's no vSAN frontend equivalent in the k8s landscape afaik.

100%. it was very hard (but not impossible) to compete against vmware in the virtualization space before the Hockening™; vSAN, vSphere and NSX were very very nice and super easy to use. nutanix succeeded more often than not.

orgs now are kind-of scrambling for alternatives. great for nutanix. better for the CSPs. also great for red hat (my employer) since we have openshift virt, which is based on kubevirt but is much easier to use and works nicely with the rest of the ecosystem.

zen? like, uh, the browser?

I assume they mean Xen.

Yeah, but I think his point is that VMs haven’t fallen as far out of vogue as the parent is perhaps suggesting.

We still run a lot of VMs, just not VMware VMs.

This really seems only obviously true if you're counting docker/podman-desktop and similar dev tools which work via stashing containers in a VM. There are a ton of large scale kubernetes deployments made directly on baremetal.

Did you switch from VMWare to Microsoft?

The bigger issue is, if you're refusing to honour a contract as a vendor, not only do you risk a lawsuit like this one. But more importantly, who is ever going to sign up for another contract with you? You just proved it isn't worth the paper it's written on.

Unwritten terms like "valid until I decide to tear it up haha lol" are not generally appreciated by companies that depend on your stuff for their business. Of course you can extort your existing customers until they manage to move away but basically in the longer term you're suiciding your entire business.

So switch to openstack or kubernetes (with kubevirt if you want VMs). Open source. Way more beautiful design.

With Kubernetes, actually fast storage if you need it. Can scale up to AI demands if you need it.

Or proxmox or the like if you're small enough.

It feels like that's the direction most people are going in, but that doesn't change the fact that no one is going to trust Broadcom again after this.

After this? Many people didn't trust Broadcom even before they bought vmware. This isn't something new on Boardcoms part, though it is high visibility and so people not even aware of Broadcom before are now.

My beef with K8s (and to be clear, it’s the leanest cut of beef from the deli - so not much substance to it) is that unless you pay someone else to manage the Control Plane for you, you’re not only going to need to upskill your workers on K8s itself but also administering the components of the Control Plane, like HA, etcd, storage, network plane, etc.

Compared to standing up literally any Linux distro and KVM, K8s remains an overly complex PITA to get off the ground and integrated into an org on the cheap/free. In that area, it handily loses to even Microsoft Hyper-V in the “just get us going” category of business adoption/velocity.

I’d really, really like to see K8s more streamlined for initial deployment than it is. It’s getting better, but I generally still have to grudgingly recommend a premium, managed control plane for any serious deployment.

For small on-prem shops that don't really want to learn about running k8s, and have under a few dozen nodes, there are definitely slim options, like Talos is basically boot to k8s, and for single app level, there's things like multi-node k0s. Tech like this means you can reduce the control plane labor and focus on the workload.

https://www.talos.dev

https://docs.k0sproject.io/v0.11.0/k0s-multi-node

Talos is on my shortlist but its core “grease” features remain locked behind a (reasonable, but still existent) subscription, which throws it into “premium management layer” territory for me and my odd slide deck for executives. The narrative for the past fifteen years has consistently been “we have no money for what we need because we spent it all on what Gartner suggested and a consultant told us we should have”, which means we’re constantly having to not only do more with less, but also rely heavily on “pre-greased” products like hypervisors.

I don’t like it, but that’s how the current technology environment is unfortunately setup.

God help the enterprise software segment if customers realize 90% of their needs are served perfectly well with KVM+QEMU and VMs.

Word for word my experience with operating k8s.

To be clear: I like K8s! It’s fun to be able to write some YAML, apply it, and be done!

But the sheer work of getting to that point, safely and securely? It ruins the experience for me, personally.

What would you like setup to look like?

Would there be real interest in a kubernetes distro that takes IPs and a (set of) domain names, and boots up on N nodes, installing letsencrypt, so that you can do a deployment and have ingress actually working?

For homelabs and SMBs, that is definitely an opportunity. Other areas I’d like to see more growth in:

* “No-Code Kubernetes”, that lets staff design basic (or even not-so-basic) deployments using a web-based GUI. We’re seeing more of this from a “understand how things work while they run” perspective, but I haven’t personally come across any “here’s your building blocks and explainers of the environment variables, go build” solutions.

* An “ESXi-ified” K8s. Talos comes so, so close to this, but I’d love something that was as easy to deploy into production as ESXi was on SD cards. Deploy as an appliance on bare metal or as a VM, and voila, Control Plane with an IP schema, network layer, AD CA/ACME support (including Let’s Encrypt), and a basic load balancer/ingress. Changing the setup is as simple as adding a basic text file with the control plane IP and join string (worker nodes), with a simple flag to add it as another control plane node for HA.

* Renewed focus on etcd management ease. A lot of the cert track focuses extensively on etcd management through kubectl, which is an unnecessary abstraction layer for things like backups, failovers, and redundancies in smaller IT departments.

* Automated migrations. The K8s evangelists hype it as being able to manage VMs, which would be great if kube-virt was standard (it’s an add-on). I’d like to see K8s either formally integrate it into the baseline or more distros make it a checkbox option at cluster creation. Part of that should also be support for automatic deployment creations for existing VMs in a hypervisor, by analyzing current settings and suggesting the YAML or JSON to replicate that VM in K8s with appropriate IP address, current storage, and ACLs

From a tech standpoint, the foundations for K8s success have long been polished into a mirror shine. At this point it’s usability and accessibility that remain broadly unaddressed, especially if we want more people and companies using iterative, composable infrastructure.

Check out Harvester: https://harvesterhci.io/

Perhaps I'm ignorant but why would you NOT have someone else manage the control plane? It's relatively easy to switch providers and there tons of options. It's also pretty cheap.

I havn't looked about but I'd be suprised if there wasn't also a large number of companies providing on prem control plane support.

Pretty sure in Tesco's case switching to openstack is a decade long project.

I think they can do it in 5 years with some investment. Which is how long they need Broadcom to honor the current contract. The effort is mostly technical, and much of it you can just hire contractors to help.

I would bite the bullet and begin any switch from Broadcom now on the principle you can never trust the vendor again and you will just be kicking the can down the road.

that doesn’t seem to be an issue in this case, since it’s exactly what everyone expects Broadcom to do in any given situation. their victims/customers are people stuck on the platform from before Broadcom bought it.

This.

I had to tell CurrentCo that I cannot reinstall their vSphere deployment at a client site because they bought a perpetual license, didn’t migrate it to Broadcom before they cut it off, and now we cannot simply go get the latest patch or appliance for that version number without inviting an audit and a sueball from Broadcom.

“Good thing Microsoft would never do that to us.”

Ha. Hahaha.

At least VMware isn't user-facing and it can be removed without riots. Imagine trying to tell someone they don't need Excel. I try to maintain at least plausible flexibility to go tell vendors to shove it, but if you have some enthusiastic fans of Microsoft Teams (they exist, who knew?)... Teams is one of those things that is inescapably tied to an incredibly deep well of platform lock-in.

Really? Teams?? We went teams abd microsoft ecosystem fully because we needed extra windows management stuff as we have grown and users had software that required windows and excel and the biggest pain point has been teams. As near as i can tell it tries to do everythibg wrong and the things that are so blindingly obvious that it can't do them wrong, it finds a way to do them suboptimally

I would never invent a lie as implausible as this. Yes, there are fans of Microsoft Teams. They're out there and they make decisions.

I... I like Teams...

So, in the spirit of intellectual curiousity, and I will avoid making any judgements in any of my responses, I have 5 questions:

1) Have you ever been exposed to alternative communicators?

2) What features do you enjoy about teams

3) What platform are you using it from (Windows Desktop / Laptop? What spec)

4) Have you ever written a bot or integration?

5) Can you take me through a very brief working day for you, with a focus on collaborating with others.. (file sharing, online chats, IRL chats, meetings?)

I'll give my own interpretation. Not that I love Teams, but the alternative in a dinosaur corporation is basically email.

1) WebEx and the open source chat that Oracle appropriated. Fortunately Zoom came and went too quickly.

2) Searching the Exchange corporate directory. BASIC features: status, embedding pictures, attaching files that Outlook would block. Sharing links that aren't obfuscated.

3) Can you even run Teams from Apple / Linux?

4) Ha! Imagine the nightmares for the person linking Atlassian and Teams.

5) Group texts, file shares, voice calls, recorded meetings. Meetings with groups from other companies is almost painless.

> 1) WebEx and the open source chat that Oracle appropriated. Fortunately Zoom came and went too quickly.

Ok, then I can see why Teams ranks among them. I would invite you to try something like Zulip or Mattermost but I think ignorance is bliss and you should avoid knowing about anything that could be better. Your mind might do this for you (rejection) but best not to tempt fate.

> 2) Searching the Exchange corporate directory. BASIC features: status, embedding pictures, attaching files that Outlook would block. Sharing links that aren't obfuscated.

Appreciate the list, the only one of these that's Teams specific is searching a corp directory. Do you use the "Teams" functionality, or do you use the chat exclusively?

> 3) Can you even run Teams from Apple / Linux?

Yes, it's very slow. It's also very slow from laptops, the best "Teams experience" I've ever seen has been in GameDev where we all ran Windows 7 on dodecacore CPUs with 128-256G of DDR4.

It was still slower than Slack on my macbook air though.

> 4) Ha! Imagine the nightmares for the person linking Atlassian and Teams.

Yeah, people do. People also use Excel from within Teams.

Writing bots for Teams is a special nightmare, but webhooks can work.

> 5) Group texts, file shares, voice calls, recorded meetings. Meetings with groups from other companies is almost painless.

Do you spend a lot of your day face-to-face or more of your day in Teams?

Do you find yourself arranging meetings to sync rather than using the chat functionality?

Do you find that people have to ask around a lot to get an answer and then ask again later when it's forgotten, or can they find their answer in history?

Is native Teams on Linux still a thing? I had it installed but the package disappeared from the MS repository. I currently use the web version.

Quite right, it seems that Teams for Linux is discontinued.

Guess this means I wont' get to run Teams in the company I'm joining, which is doing all its security attestation via Microsoft;

https://news.ycombinator.com/item?id=32678839

The web version runs fine in a chromium based browser.

It also refuses to run well in non-Chromium-based browsers.

Yet more vendor lock-in.

Yes even when I log into Outlook M365 from Firefox on Linux it's constantly kicking me out. Every time I get back to my computer it gives me this passive agressive "Please hold on while we're signing you out..." when I never asked to sign out. Grrrr.

I also notice it works way better in firefox if you set the user agent to Edge on windows. Some features which are broken in firefox suddenly work totally fine. Which reinforces my belief that they are simply doing this to bully people into using Edge.

There is a PWA you can install, with it's own icon and everything. Yes, it's a not-even-glorified-web-browser, but meh, it works (for some definition of works).

There is a native Teams package in the AUR which worked well when I had to use it. I assume it will get outdated eventually from lack of MS support. Web works of course.

Someone patiently explained and introduced the "Teams" feature of "Teams" to me. It's easy to ignore. Here's a tough one: ever used a Microsoft Loop component?

My preference is text chat but we do a lot of unscheduled voice chats when screen-sharing is involved. In-person meetings are nice when possible, it's been easy enough to connect a Teams meeting from a conference room phone.

Before Teams I set up a Mattermost instance, and I think RocketChat integrated to GitLab? Nobody used those. As we all know the value in these things comes from network effects; with Teams corporate IT can set Teams as a startup app by Domain policy, now everyone in your company has to be online. That's the real killer feature.

I find that MS Teams works pretty well if you don't put too many channels in a team, or at least don't use the extra features in the sub-channels. Nothing like having wiki/files spread across multiple channels under a team as opposed to a directory structure.

To me, what I really don't like is that you can present the files and wiki via network file share, but the format of the wiki is read-only in that mode... it would be significantly better as user editable markdown with front-matter. It's some strange quasi-html extensions that you aren't going to be able to really even use.

I don't like that they separated the chat channels out... you now have unread, channels, chats, meeting chats... going between them is now a mess as I'm usually in a combination of meeting chats and 1:1 chats mostly.

While I do appreciate the integration of outlook's calendar, I really don't want meeting notifications for anything more than a couple days out... it can wait until I check my mail.

The real time video chat is okay... I think Zoom and Slack both have lightly better video quality and features though. I don't like Zoom's chat/discussion features nearly as much... I like Slack overall slightly better, but don't think some of the externalized integrations are as good as Teams.

Overall, it's "okay"... I don't love it as I think it's gotten worse over time while Slack is getting better, slowly. I'm not vocal enough to raise hell one way or another for/against it. I've used it on Linux without much issue... so it works well enough.

> 3) Can you even run Teams from Apple

Unfortunately. Teams is just as performant on MacOS/iOS as it is on Windows.

Teams is a React App. Teams classic uses Electron - so perf was identical. The New Teams uses native platform web-view, so mileage may slightly vary. Still a React App, though. sighs.

Teams works better in a browser window in Linux where if it hogs too much CPU Cromium pulls the plug.

That's how I run it.

> Fortunately Zoom came and went too quickly.

I've used both Teams and Zoom (and others). Honestly, I'd rather use Zoom instead of Teams.

> BASIC features: status, embedding pictures, attaching files that Outlook would block. Sharing links that aren't obfuscated.

Status is settable by just about any competitor to Teams. Slack and Zoom both can set your current status.

Embedding pictures and files is also not unique to Teams.

Obfuscated links? Just a matter of time before Microsoft changes that to some microsoft link for a "vulnerability scanner" and then charges the company for the privilege to block random things it doesn't understand how to scan.

> Can you even run Teams from Apple / Linux?

Yes / technically yes (not supported any more)

> Group texts, file shares, voice calls, recorded meetings. Meetings with groups from other companies is almost painless.

Slack and Zoom are better at all of these.

Zoom FTW, big-time.

I wouldn’t say Zoom is better to be honest with you, for just meetings the UX of Teams is pretty bad but the UX of Zoom is almost as bad; there’s not much in it.

Last time I checked Zoom was a pig on resources and required a weird background worker- and you couldn't even send files.

> Can you even run Teams from Apple / Linux?

I can tell you I've successfully been in Teams video conferences from my Linux desktop, in the browser. I was surprised too.

Teams is fine, especially as others are so expensive for small non-US shops

We already have to bite the bullet and pay for office, at least we get free chat

I wish Teams integrated better with Github Issues/PR, but it works well as a company-wide chat

Yeah, it's fine. It's apps in a browser. It basically works. It's as good as anything else I've tried.

Teams brings back a sense of adventure into boring online meetings since you never know what works subtly different than id did last week and who will be made to act the clown due to strange glitches.

Poor mac users.

The only way you could be a fan of teams is if you've only ever had to use stuff like cisco connect or lotus notes chat. It really is just terrible. Teams on my window laptop makes the fan go more than running a pretty massive compute cluster, its crazy how non performant it is.

Something is misconfigured on your machine or you have corporate spyware like Carbon Black driving up processor load.

Theres definitely corpo spyware. Id never use windows on my own machines. Using windows and wsl is just path of least resistance right now.

Maybe, but teams is really, really, really non-performant. Its actually kind of impressive of dog shit slow it is.

To be fair, zoom is somewhat slow too. Not to start up, but joining a meeting is like 5-10 second delay. Lord help you if you turn on your webcam.

You didn't really say much of what does it do wrong or right, you seem to just try to convey a for-granted idea.

I've been an msft employee for a couple of years and teams... Was ok. I prefer slack, but meetings, video, messaging, formatting, etc. was just fine in teams.

i agree. teams is "ok", for my purposes at least. don't really understand the hate it gets in certain forums

There's a bunch of annoyances in teams. The text editor is awful if you try to paste/send code for one. The screen sharing is shit by default and only lets one person share at a time for no good reason. It's fairly slow. And it doesn't have a global fucking hot key for push to talk/muting. Despite being integrated with the damn task bar(which is broken if you're on hold in one call and active in another, then it's too stupid to figure out that the mute button is intended to mute my active call, not the one that's on hold...). Also the teams feature gets out of your way, which is an issue if you're trying to use it like you would channels in discord or slack. Would be nice to at least have a notification pip there.

> And it doesn't have a global fucking hot key for push to talk/muting.

It's control-space. It is global on Windows as far as I know. Unfortunately you can't change it because I would love to have a single key I don't normally use assigned to it. I use a Mac keyboard with 19 function keys so there's plenty that I would never touch.

I agree with all the other criticism by the way, it's a messy slow clusterf*k.

I hate to be like this, but no it's not(at least not on my windows 10/11 work laptop). It is not active globally, the Teams window containing the call has to be active(on my system).

And it would be a problem if it was, as ctrl shift is the default binding for show suggestions in IntelliJ and Rider.

All of the following you've already mentioned but I forgot you mentioning them in my rage at MS. Figuring out if I'm holding it wrong or not revealed another silly issue however: the push to talk button is hard coded to ctrl space, unlike other keybinds it doesn't show up in the keybinds dialog.

Messaging is not fine. There are 15 million group DMs and zero way to organize them outside of a giant list and a favorites. That's where Slack shines.

The text editor is hilariously broken. Copy/paste eats formatting for breakfast... Unless it's Word, but only sometimes. But who the fuck is pasting a WORD document into teams?? Don't do that!

Seriously, it's outlook levels of broken. Markdown doesn't work, I don't know what markdown engine they use but it's certainly not compliant to any sort of standard. Copying whitespace is just unbelievably fucked, your code becomes unreadable.

Which would be fine... If we could bulk indent. But do you know what highlight + tab does? Not indent, no, it selects the send button.

Teams, fuck you.

I’ve got users rioting over the fact that we might remove Mattermost and move them to Skype/Teams. Note that I am in airgap and can’t use Slack. I am looking at Rocket.Chat though since MM is $$$$$!

Controversial choice (UX isn’t super pretty) but I recommend Zulip wholeheartedly.

> someone they don't need Excel

well I have news for you ;=)

I pity the people who have to deliver news like that from time to time.

You don't need Excel. Have not used it in over 10 years.

Teams, that application IT is forcing me to use because they are a "Microsoft" house. Same application currently stating I'm on the _Calendar_ screen on the Task bar but actually in the _Chat_ screen; _Calendar_ bloatware feature and others have been removed and will always. Even when _Microsoft_ screws the user and force a reinstall of features after a Teams update.

Microsoft is a trillion dollar company that rejects quality user experience, QA, and is great at producing crap-ware. There is not a single product sell that I will spend a penny on. Still waiting on that 7+ year request to destroy | delete dangling pull request in Azure.

> You don't need Excel

This is the kind of thing that gets tech people a bad reputation. YOU don't need Excel. I don't need excel - but that guy? You have no idea what he needs and if the people he's supporting need (or just want) Excel to get their jobs done it is incredibly arrogant to tell him what he does or doesn't need.

Now, I've loathed Microsoft since the 90s, but that makes me a weird and special little petal - it doesn't count for squat in business.

Right, and even "that guy" might not need Excel but the second he opens up Libreoffice or Google Sheets and something doesn't work the way he's used to, he will say it's broken. He's not interested in learning how LibreOffice or Google does it, he's just trying to close a $5m deal. And he's not wrong.

Honestly, these alternatives just don't stack up against Excel. Even setting aside the advanced stuff like complex data analysis and macros that some organizations rely on, Excel is simply more robust and user-friendly. Google Sheets feels like a toy in comparison.

And a toy is all most users need. BUT (as has been the case with me at various points in my career), some need a lot more.

And uhh you don’t need visual studio or pycharm. Just open notepad and learn how it works. It writes code.

Indeed, or in my case where I don't need Excel, but can't tell someone else they don't. Telling me in this discussion I don't need Excel is very much misreading the conversation.

I never said that guy does not need Excel. High probability that all the features in Excel being utilized by most users is available else where. I get told all the time you don't need TOOL_X because that tool is uncommon compared to office software. It is funny how many people think everyone uses spread sheets. I personal don't need them or touch them. There are others tools that are better fit.

Excel is a great definition of _Law of small numbers_ when it comes to praising Microsoft. Major of all Microsoft solutions are not great and there are better alternatives. Even the most popular ones where not created by Microsoft and bought.

Microsoft does not have any user experience standards. This was ingrained with Bill Gates making CTRL+F _Forward Email_ in Outlook versus a constant of _Find Text_. Number one complaint I see from Microsoft employees in their feature tickets is lack of _muscle memory_ designs. Their file listing algorithms still think 109 < 74 where quality file listing should show 109 after 74, not before.

Microsoft produce products _You have to pay me to use_. I have not bought a Microsoft product in years and never will with my personal income. This doesn't not mean I don't know how to use their junk-ware -- that is more tools in the tool box. It is the mentality that once you know one spread sheet software you don't need any other. Once you know how to use Microsoft tool X, you don't need to know how to use tool Y, even when tool Y is a better fit for the solution.

Then again most people like to short their brain and only know the tools they were taught or used in school.

> I never said that guy does not need Excel.

You literally wrote "You don't need Excel."

Imagine an accounting person telling you: You don't need version control, have not used it in 10 years. This is a _shockingly_ arrogant and ignorant comment and if you actually believe this I fear for any organization in which you have significant power

Some of us deal with the 95% of users who don't give a steaming crap what some condescending IT droog thinks about Microsoft. Excel is not optional out here in the real world with the unwashed masses.

Isn't their whole strategy that they want to squeeze the customers they got by the balls?

https://www.youtube.com/watch?v=WpE_xMRiCLE

Mostly they seem committed to drop smaller customers and pursue very lucrative deals with large companies only. But to do that they should be valuing those relationships with the big customers, and they clearly aren't doing that either.

Ops person here, VMware is so embedded at these companies that switching away would be like Google saying, no more gRPC, everything is now SOAP. The amount of impacts is just too mind boggling to even consider.

2 companies ago was heavily invested in VMware. It impacted monitoring, backups, deployments, networking, cloud migration and more. I can only shudder at level of effort they might be going through to get off VMware.

Because of that, they probably won’t for years even as Broadcom screws them over.

The bias with technical people is thinking every problem is technical

VMWare may have hiked the prices and might be an important dependency but at a certain point it is cheaper to sue and/or switch from them.

Seems that they have gone way past this point

Switching away today is too mind boggling to consider, but switching is purely a technical problem. We can put a price on the costs, and the time. VMWare was founded in 1998 - that means 27 year ago nobody was using it, and in turn we can say took you less than 27 years to get dependent on and - surely you can switch to something else in 27 years. More likely you can switch in 2 years - that is about what it took one company I know of.

Sure but old company cut Ops team to the bone, which is why I don’t work there anymore. So CTO is faced with, pay VMware or cut back on deployments so Ops team can have some breathing room to work on migration to whatever they pick.

... Google did say "No more Oracle EBS" and switched entirely to SAP. It took multiple years, and it was not a small effort, but there was the will, and a way was found.

sigh I didn’t work at FAANG type, we don’t hire FAANG skill level. Whatever Google did is completely irrelevant to this conversation and most other conversations.

In my experience, FAANG are no more skilled than anyone else. The main thing is that they seem to do a better job of not hiring complete duds, so their average cleverness may be higher, but I’ve worked with brilliant people at every regular shop.

I agree but complete duds can really drag your team down as you end up spending a ton of time trying to fix their shortcomings and amount of code written on the Ops side to prevent them from completely screwing everything up is mind boggling.

AI-assisted migration of glue boilerplate code transforms this mind-boggling amount of impact into a two-year project, max.

Only to the extent that the team is competent enough to properly test all that boilerplate code, which is very far from a given. A relative of mine in IT has had a large internal-tooling migration get dragged on for years by the persistently bug-ridden code of one of the groups working on it.

These days it is really hard to figure out if comments like these are real or satire.

The AI can’t migrate the knowledge of all the people that have to operate your on-prem VM deployments.

What code? An on-prem VMWare deployment is all about hardware, storage, networking, and fuck-tons of planning, budgeting, and approvals. There is little to no customer-written code in a typical VMWare farm, except maybe some Ansible or whatever for minor customizations and automation.

The bigger issue is the lawyers get the money so go back to school as a lawyer and choose better parents.

As someone who only knows Broadcom's silicon business: there, they're just used to people having no other choice, with their quasi monopoly in some fields. Are they (mistakenly) transferring that attitude to VMware?

Yes.

Our 5 year ELA for vmware went from 1.5M USD to 12M USD. Higher ed.

Our Hyper-V environment is coming online this month. It was already included with our ELA with Microsoft so we were able to splash out a bit for some higher tier support.

I’m curious, why does higher ed need VMWare? Is it because you have some supercomputers that you want to share among your employees?

I’m asking because for $12m I’d just buy all the employees some high end commodity hardware.

It will be for ordinary administration, employees, students, finance etc.

The academic work (including the supercomputer) will run Linux and open source systems for job scheduling etc.

Why does any of this use VMs? Not familiar with the space at all

Uh? IT for higher ed is not that different from any other enterprise IT.

I used to work at a university, although I worked on the Linux/HPC side of things we did have regular contact with the IT department. So things the IT department used VM's for of the top of my head:

- Web servers. Yes, the official university web pages, with fronted servers, database servers and whatnot. But also a lot of departments had their own servers, even some research groups ran their own. To get rid of the zillions of ad-hoc servers running in closets here and there IT gave out VM's pretty freely to staff members.

- Email. Yes, this was before everyone + dog outsourced their email, so they ran their own in-house email servers.

- print servers

- (I think file servers were mostly non-VM appliances, my university used netapp's a lot)

- All kinds of management systems to manage the campus workstations and network. And things like Active Directory and other directory services type services which are critical.

- A zillion in-house applications for things like signing up for courses and other things necessary for handling thousands of students.

- A lot of bespoke systems given out to research groups for whatever purposes they needed, again in order to get rid of the zillion repurposed old pc's running in closets acting as servers or running some experiments etc.

- Critical services and some not-so-critical services as well had test environments to test changes before rolling out to production.

- Finance/admin stuff like payroll etc.

- Shell servers (ssh), RDP servers, VPN servers etc. to enable staff to access university services from outside.

All in all, it was hundreds and hundreds of VM's. Wouldn't surprise me if there were actually thousands.

None of it needs to be on VMs, but it's generally more convenient to manage when it is. You could also use something like Kubernetes, but then you're administering Kubernetes.

Higher Ed is really just Enterprise IT at a giant non-profit, sometimes state funded, sometimes not.

The HPC side of the house actually used VMWare _less_ than the enterprise side. Mostly due to funding restrictions.

Mistakenly? No, Broadcom was very up-front about their plans to offload small customers and massively upcharge large customers pretty much the same week the purchase was announced. It's stupid, and Broadcom are certainly assholes, but they did give a LOT of advance warning.

> who is ever going to sign up for another contract with [VMware]?

Oh geez I dunno, surely now it's over for them

There seems to be a "squeeze" phase in several of the software's life. When the organization goes from engineering oriented into sales oriented. Mostly true for those acquired by investors, wanting more money on spending no more. EBITDA and other related matters becoming paramount having no decade long perspectives, not even half in some instances. No long term there. Of course, you have to find the gold mines, those with locked in users.

I have never understood how Oracle remains a big company. Seems to me they have this type of reputation and just nobody will ever willingly buy anything from them ever

The state of software companies is pretty terrible. I have been on the acquisition side as well as the development / end-user side and it’s mind-boggling knowing the exorbitant costs with bare minimum value delivered, yet companies just keep paying whatever they’re told it costs, until it’s comically astronomical and the customers have to tell them to get bent. Yet still, software vendors keep changing their licensing structure until it meets that comically astronomical figure and pushing customers away.

Enterprise software licensing, support contracts, and technical account managers (TAMs) often run into hundreds of thousands or millions annually per organization. Yet, in practice, support tickets go unresolved or ignored, even for large clients.

The software quality of our most expensive products is extremely poor and unreliable, almost across the board. Many products suffer from bugs, outdated features, or incompatibility issues that disrupt operations. In development roles, this means wasted time on workarounds, custom patches, or integrations that shouldn't be necessary. For a non-small organization, this scales up to significant productivity losses and hidden costs in overhead.

These companies actively alienate us, the customer, through their business practices. Changes like aggressive licensing shifts (e.g., moving from per-core to per-employee models) force reevaluations and migrations and eroding trust (i.e. Oracle with Java, VMWare fiasco). This isn't isolated—it's a pattern where short-term revenue grabs risk long-term relationships, yet companies seem unfazed.

This jacks the entire ecosystem up. These practices stifle innovation by locking customers into suboptimal tools, increase overall IT spend industry-wide, and contribute to employee burnout in dev and ops teams.

It seems like it’s a race to the bottom. The strategy is to create an ecosystem with high switching costs and vendor lock-in. It just doesn’t seem sustainable, yet- it keeps truckin’ along.

Charlie Munger once said something to the effect of "show me the incentives, and I'll show you the outcomes".

There is very little incentive to produce high-quality software, to not alienate your customers, and to support the thing you already sold. Those things cost money. Money paid for those things is money not paid to shareholders, and that's the ultimate incentive in our system.

They've got you by the balls, and secretly, your CEO thinks their CEO is a genius for thinking up and implementing that business model. Pay up.

> There is very little incentive to produce high-quality software, to not alienate your customers, and to support the thing you already sold.

I think some of what I'm trying to portray is that this should be the incentive. Either do it or you don't have a customer. Yet, the customers don't hold this standard.

As an individual consumer, if I don't get what I pay for, I return it or can even submit a charge-back. Is it not irresponsible of business management to not do the same?

Software, done right, is both extremely hard and expensive. Hardware was cheapened by China/Asia but it is not happening for software (theirs generally sucks and they lack many fundamentals). Europe completely lost the race.

The current breed of managers in the US have decided to fire developers, abuse customers (you have nowhere to go) and burn all the money on AI (they believe it’ll solve all their problems).

Morale will remain low until an alternative spawns. Kinda with electric cars. Europeans, Japanese and Koreans are now forced to up their game and lower their prices.

You're just saying things you want to be true, "Asian software" doesn't suck and Europe didn't lose.

Just because "all" software companies have American entities doesn't mean you "won", that's just what happens when a jurisdiction let's companies do anything even if it's detrimental to society as a whole.

Most (all?) of the tech stack depends on American companies starting from Operating Systems, to Servers, to SaaS, to Cloud, to Software running on most businesses, etc. What you are saying is meaningless when they got you by the ba&&s.

Yes American companies have been good at capitalizing on IT and a lot of companies are "by the balls" of Microsoft, but much infra is opensource. Linux runs the world.

Linus moved to the US two decades ago, so it's been quite a while since Linux could be considered "European".

I think it should be considered what it is, global with profits centered in the US.

Certainly, the are a fair number of contributors in Europe and both SUSE and Canonical are European based. I still think it's hard not to think of Linux as fairly US-centric however if only because of how many large US companies use (and contribute to) Linux.

Valid, I've noticed that a lot of non-US contributors work for American corporations too. It's a bit ideological for me ("OSS is global") so I think I might gaslight myself into believing it's more equal than it is.

I'm happy wherever the contributions come from either way but I will never call Linux US-centric!

Lennart Poettering(German, works for American corps) comes to mind as an example, though not a kernel developer.

Most people do not interface with Linux on a bare metal, self-assembled server. They use AWS, GCP and Azure. There is Alibaba Cloud but it is so bad, I can't even properly signup/signin.

> Linux runs the world.

The world infra runs on top of Linux. Linux is open source. Most of this infra is American.

Is AWS Stockholm American? It's a bit of a stretch, it's profits surely go to America and the control plane is American.

Another way to think of it: AWS Stockholm is a pipeline to navigate Sweden regulation/laws/banking to funnel money out of there. The core is still American/controlled in America. Otherwise, people would have just hosted in a Swedish hosting platform.

Swedish power, Swedish backbone, Taiwanese chips and boards, and an American control plane.

But yes you're right, it's an American service offering that I for cost reasons and I avoid big cloud like the plague. USA surely knows how to charge for their services and lock customers in

It seems like these problems are related to software and a certain size of corporation that is selling it. VMWare isn't hoping to make a few $million, they want/need to make a few $billion. And it is absolutely a race to the bottom when you can get Proxmox for free. Nevermind Harvester and some of the other projects out there are are doing kinds of similar things.

I was a beta tester for VMWare way back when. It was one of the first pieces of software I bought out of college. It was like manna from heaven, I could commit to Linux and have a backdoor for Windows and I needed it, and I did from time to time. I also did security testing over the years and once you've joined a machine to a Windows domain, it never can be made the same again. Vmware enabled that business without a spare laptop or spending tons of time rebuilding it. I've maintained the license since then, 25ish years. Bought it for the Mac too. I can't think of a worse transition than the one they're doing.

My pet peeve is core-based licensing for products such as database engines. For that matter, any kind of capacity licensing tied to some variant of Moore’s law inevitably results in the vendor holding their product’s face under water as the tide rises around them.

As a random example, SQL Server Standard Edition is limited to a “very generous” maximum of four sockets, 24 cores, or 128 GB of memory.

That’s just slightly bigger then a laptop these days!

Azure offers a new VM series where the max memory limit of SQL Std is exceeded with just four cores (8 vCPUs): https://learn.microsoft.com/en-us/azure/virtual-machines/siz...

There are VMs available now that have crossed the “kilo core” threshold. You can draw pictures in their task manager by creatively putting load on the processors: https://learn.microsoft.com/en-us/azure/virtual-machines/siz...

The problem here is that Microsoft kept their license limits as constants relative to a reality that moved exponentially. They would have to have applied “inflation”, but they just saw their sales figures go up and up… and nobody will rock that boat!

Inevitably they’ll keep choking their product until it turns purple and dies. It’s a force of nature, there is nothing anybody can do do counter this naked corporate greed that is enabled by accidental mis-pricing. This can never be corrected, except by letting products die and be replaced wholesale in the market.

Time to learn PostgreSQL, I guess…

Agreed. My org can take some of the blame but with our big vendors:

10s of millions on Adobe - admin side craps out all the time, terrible outsourced support (easy to escalate past them but still annoying that we sometimes have to start there)

10s of millions on Microsoft - Laissez-faire attitude about when they will update stuff during renewals and true-ups. Not specific date. Little help.

Broadcom - PE play (as we know from Broadcom) to squeeze every penny out. Have been through several account managers now. They won't sell through VARs anymore, only direct, and to get anything beyond VCF you have to be on a very special list that supposedly the CEO personally approves (which is beyond mind-boggling).

Starting to see some post production software be licensed based on having limited amount of contiguous time zones which is also crazy for a global company

Note: Media and healthcare industries may be SOL because there are lots of content and cybersec 'requirements' for having private clouds. Curious what others in those industries are doing

A lot of it is the cost of doing business. We'll see how it goes when MS moves to the consumption model they're proposing.

I know someone (I'm not going to say who - too many sue-happy people are accused here) who was using VMWare at work, and they decided to switch to virtualbox because they trusted Oracle more than Broadcom. Oracle has long has a reputation of being licensing jerks, but they are still trusted more than Boardcom.

Long-term VMware customer here and I am in the same boat as your friend.

Broadcom is headquartered in Palo Alto, and their physical product lines are mostly manufactured outside China from what I can tell (Broadcom itself is fabless). Are you sure you don’t have them confused with another company?

Yes, sorry, my fault. It was indeed another company, I have updated the original message.

This is the ghost of Charlie Wang haunting the software industry. Computer Associates was notorious for this kind of licensing shenanigans. Guess where Computer Associates is now? A new generation of IT departments are discovering the Long Island wiseguy approach to licensing.

We ran 100% of our workloads on VMWare this time last year. We'll be at 0% this time next year. We were heading that direction over the long term anyway, but the Broadcom shenanigans made us double down on that effort. They may actually be more unpleasant to deal with than Oracle, which is something I would have thought to be impossible.

In a past life in the mid/late 2000s, I did 10GbE NIC drivers for a small IHV. VMWare was by far the most awful vendor to deal with. They had mandatory certification testing which was required to distribute the driver. Their tests were so much worse than MS WHQL. There was invariably something broken in their tests that we had to work around. Each time this happened, we had to go through their support (And pay for the privilege) to tell them their tests were broken and to give them patches to fix it. This would happen pretty much every driver release, and we would end up dealing with a different person each time.

My favorite thing about leaving that job was never having to deal with VMWare ever again.

https://www.forbes.com/sites/stevemcdowell/2025/08/31/broadc...

> Many mid-market and regional operators view the new [subscription] structure as untenable and are actively exploring alternatives.. Nutanix emerged early as the leading competitive alternative to VMware.. over 2,700 new customers.. driven by organizations fleeing VMware's new pricing model.. [including] more than 50 Global 2000 companies, representing major enterprises willing to undertake complex, multi-year infrastructure overhauls.. With VMware serving approximately 200,000 customers globally, Nutanix sees most of the migration opportunity still ahead.

> Many mid-market and regional operators

These aren't Broadcom's ICPs.

> Nutanix

Good for Nutanix. Market segmentation exists for a reason.

The article is also written by Steve McDowell, who's analyst firm (NAND Research) is sponsored by Nutanix [0][1]

Welcome to Enterprise Sales.

[0] - https://m.youtube.com/watch?v=93FbVZGDXoY

[1] - https://www.nutanix.com/theforecastbynutanix/technology/hype...

The article mentions several VMware alternatives:

  - RedHat OpenShift (k8s)
  - Scale Computing HC3
  - Wind River Cloud 
  - public cloud providers

Any other alternatives?

Proxmox leaps to mind. HyperV, XCP-ng, raw KVM..

Xen is still around, including an open-source version. It's not as visible as it used to be but works e.g. XenServer (https://www.xenserver.com/editions). I'd look at this over VMWare anyday.

When I was looking at solutions, most of what I found seemed to indicate Xen is waning in popularity. I considered XCP-ng but since KVM seems to be more preferred now, I ended up going with Proxmox for a few small work (3-5 hosts) and home (1 and 3 host) systems. It’s actually been rock-solid, basically had zero problems with it.

> rock-solid

Perl FTW, https://github.com/proxmox/pve-common

If you're looking at small self hosting then Proxmox.

That’s where my IT department said they’re headed the other day.

in no particular order:

- HPE morpheus VM Essential

- Apache Cloudstack

- Platform9

...

Really feels like Openstack could use a fresh coat of paint and someone to create a nice opinionated distribution of it and challenge some of these f500 customer vmware accounts. It's very powerful, but is as user-hostile as I have personally experienced. Maybe this already exists though with Cloudstack?

Proxmox is an amazing alternative

To Broadcom you’re not a customer, you’re a mark, a patsy, stooge, a _victim_. Their aim is to establish exactly what they can get away with, how far they can abuse you, before you’ll just walk away.

But this is where all/most “platforms” go. As the product offering flounders over time, your quality talent (engineering and business) boils off to other opportunities. Then the short term value extraction methodologies show up, and everyone looks on in horror as the platform is “destroyed” through “mismanaged” consumer relationships.

Working in agtech, I’ve always wondered if this isn’t just the disenfranchised farmer story.

Give a farmer 1,000 acres to farm, and if they’re playing the long game, they’ll intermix their high value crops with responsible crop rotations. Managed well, this business can go on indefinitely.

But tell them they have 5 years left to farm the ground, and that the land will be of no value after that, they’ll grow the most expensive crop they can every year, soil quality be damned. It makes the most sense from a value extraction point of view.

Broadcom seems to be the kind of farmers that buy up forsaken land and extract as much value as possible before it finally fails.

Remember: Broadcoms’ (the non-chip part of it) business model is to buy tech companies in long term decline (VMware: check) that have a significant locked in customer base (VMware: check), slash development and support while raising prices.

They extract as much cash out of the decaying corpse as they can, and then discard it.

They were actually fairly open about this to investors when they bought VMware .

That made me wonder why I haven't heard of Computer Associates lately. Turns out they're part of Broadcom now.

"Sounds like CA" is exactly what I thought as well. And now I remember why.

> As The Register's European editor wearily remarked: "Search the site for Simon and VMware. We've got pages of this stuff. Go. Look."

In case anyone else was wondering who "Simon" is, I'm pretty sure it refers to this editor: https://www.theregister.com/Author/Simon-Sharwood

TIL VMware Fusion is free now. https://blogs.vmware.com/cloud-foundation/2024/11/11/vmware-...

"As of March 2025, the current versions of VMware Fusion Pro and Workstation Pro are available at no charge for all use cases, including personal, educational, and commercial use. Users no longer need to purchase a license for these versions for any scenario. Broadcom will continue releasing updates and security patches but will no longer sell new support contracts for these versions."

FYI. I know we weren't the only ones left gun shy after Oracle's legal team reached out to let us know about the dozen or so employees using VirtualBox from their work laptops a few years ago.

Yes, but just try to download it from them. You have to make an account, agree to all kinds of garbage, and then you still might be stuck in "authorization" like my account is. They removed the auto-update functionality (even the notification!) from the app, so you have to manually go through their horrendous website to download updates.

It's 10x easier to just search reddit or internet archive for a link to the .dmg.

If your name or email address has non-alphanumeric characters, their entire authorization/EULA flow breaks silently. I had to use an old email to get it to work.

> They removed the auto-update functionality (even the notification!) from the app

Oh, they didn’t want to deal with the “token” bullshit they added to the VMware download site that the updates come from. That’s why.

https://blogs.vmware.com/cloud-foundation/2025/03/24/downloa...

> They removed the auto-update functionality (even the notification!) from the app, so you have to manually go through their horrendous website to download updates.

That's not true. The auto-update functionality is still there, it just doesn't find anything. (At least this is the case for Workstation; I assume it's similar for Fusion)

Wireshark it and watch the 403 kick back from the server (because of the new token policy).

Yep! I use it to run SolidWorks on my M1 Macbook. (They sure don't make it easy, though - one of their many EULA approval steps fails silently if you have any non-alphanuneric characters in your name or email address. Feels par for the course with bigcorp software that assumes you'll have an IT team to deal with things.)

Ok, but Parallels Desktop is actively supported.

Kind of funny to see businesses screwed over by "lifetime" deals.

Most startups struggle to get any clients and would lavish love on any the could.

Imagine being a company so big that your strategy is to kick your clients in the teeth then throttle revenue out of them.

You can either buy our product voluntarily or we can make you pay for it anyway.

VMware Fusion's graphics drivers suck harder than a Dyson. For the love of god, I just want to play some games without rebooting to a native Windows environment. I'm not even asking to run, say, Crysis, or something... I guess my days are numbered as an x64 Mac user but let me at least go out with some dignity.

Perpetual licenses are eventually very bad for business if nobody pays for support. Nobody pays for support if they remotely think they will not need it. Shit starts hitting the fan when you actually need excellent support and get a (huge) invoice.

People would care a lot less if Broadcom had very gradually increase prices over 5 years or a decade, stopped support on version 7, stopped development on version 8 and gradually changed everything starting with version 9, but they decided in all their wisdom they wanted their investment back ASAP instead of waiting.

Maybe they should have looked at the licensing and support models Veeam uses.

Nobody pays for support if they remotely think they will not need it.

Maybe in your experience. Virtually every company I've worked for or with in the last 2-3 decades (mostly enterprise shops to be fair) paid for support on critial infrastructure like VMware. Most of them are required to by various contract or regulatory requirement (like the Very Large Financial I'm at now). The tiny number of shops I worked with that blew off support were 1) SMB, and 2) run by idiots. YMMV.

Agree. It's my experience and most of those are small businesses.

As soon any certification is required for the business the chapter SLA (service level agreements) comes up. Basically every business that is medium to large has to pass some kind of certification. Most of them will pay for an annual support contract because the SLA chapter forces them to have some kind of measure to recover from architecture problems.

Play stupid games, win stupid prizes.

If you need 40,000 servers to keep your business running (which you don't, your ~3-8 million weekly transactions can be processed on 1 computer, but whatever), hire people that will work on you, and whose paycheck depends on keeping those computers working, to keep those computers working.

Game theory arguments like "they wouldn't screw me over because other people won't want to do business with them" don't work when the other party is trying to maximize quarterly earnings, and their long-term thinking is in the order of ~2 years.

To be fair, and I know nothing about Tesco’s actual stack, a large grocery chain needs to track their contracts with suppliers, track their inventory in each location and in transit, track what goods they want in which locations, understand which larger pallets and big boxes contain which goods, track things prepped in house, and also optimize what to move from where, to where, and when and how. The latter part probably uses some spiffy stack involving something like CPLEX or Gurobi, and it’s not running on their “1 computer” OLTP stack.

That being said, I don’t see what 40k servers is for unless the POS machines are thin clients that use a substantial fraction of a server each.

A long and storied history of nasty surprises has taught me to never underestimate the complexity of an unfamiliar domain.

In that sense I'm surprised it's only 40,000.

Yeah, I'm pessimistically sure that there is other stuff, like:

* Checking whether each item scanned has satisfied a logical contract for a discount, some of which may be per-region, per-store, or even per-customer.

* If multiple exclusive coupons or deals are available, resolve the contradiction, preferably in favor of the customer.

* Check if any items or quantities of items require an ID to be shown before proceeding, and record information about the employee authorizing it.

* Update customer "rewards" data and generate any special offers so that you can put it onto their receipt.

And that's not even starting to get into all the other less-customer-synchronous stuff that you still need CPU power somewhere to do. Managing stock levels, orders, deliveries, price changes, anti-"shrinkage", employee shifts, market-research, status and repairs of freezer-units, operational logging and telemetry, every form of reporting/dashboard "strategic insight" stuff beloved by upper management...

If you're doing 10 million transactions per week (which is likely way more than what they're pulling) that's about 16 transactions processed per second. You can add inventory management, payroll management, you can run the company's email server, write all that in JavaScript, and you'll still have room to run a Minecraft server on the same laptop.

My point was not that running all that on one computer is a great idea, just that 40,000 servers for a CRUD application is way past what should be considered reasonable.

But even that's fine. I like computers, you can have 40,000 of them if you want, even if the only reason they exist is some guy's job security. However, you're insane if the guy keeping them running doesn't work for you.

You're suggesting that all Tesco branches should do all the PoS transactions online to a single server?

Tell me you've never designed a system at this scale without telling me you've never designed something at this scale...

Running it on a single laptop might be an exaggeration, but I can't imagine there's any essential complexity that requires more than a few dozen servers.

No, I did not suggest that, in fact in the very comment you're replying to I said:

  My point was not that running all that on one computer is a great idea...

Regardless, if you want to strawman my passing remark, I'm happy to defend it.

Let's even say my numbers are wildly wrong, and they're processing 100x more transactions than what I claimed (which was already an overestimate). Tell me why you can't process 1600 transactions per second on one computer, especially for a country the size of the UK, where you would expect a ~15ms ping when talking to a server on the other side of the country.

It seems that you should work for Tesco, they would probably pay you well for reducing 40k VMs to 3.

I would expect outages when talking to a server on the other side of the country, and an outage preventing supermarket customers from checking out or supermarket staff from usefully restocking shelves would be a very expensive mistake.

A good system here will be distributed.

That being said, two servers (for redundancy) physically located at each branch ought to do the trick. Tesco has a bit over 5000 branches [0], so that’s 10k of the 40k VMWare seats right there. Throw in some extra seats so storage and compute can be separated at each location (maybe unnecessary but comes with some benefits) and so that there is still redundancy while a server or two at a site is being re-imaged and 40k seems about in the right ballpark even for a fairly lean implementation.

And, sure, all those on-site servers might be relatively inexpensive industrial units designed to tolerate a toasty, dusty, and occasionally damp closet that looks nothing like a tidy datacenter, but it still makes sense to run something like VMWare on them.

[0] https://www.tescoplc.com/investors/reports-results-and-prese...

Most admins are more keen to shift blame than to keep things running. Having another company to point fingers at is more attractive than a proper functioning team.

Numbers I found vary but Tesco has around 3500 stores in the UK alone alongside other chains they have a hand in. They also have a large online presence, click and collect operations, estates, data collection schemes and a whole logistics network to operate. I'd have actually thought it would be higher than ~11 VMs per store.

That works out to ~11 computers/licenses per store, which sounds a tad high but also very easy to do if you let new system accrete over time and factor in the need for offline operations and redundancy across regions.

The cashier register is probably running on a vmware.

I'd hate to be the lowly, underpaid sysadmin who responded "40,000 servers" when asked the current number of servers, but he meant to respond "4,000 servers". LOL

You are severely underappreciating how complex a retail organization of Tesco's size is.

I work in this space for a retailer almost the same size as Tesco and when factoring in all the attendant organizations, businesses, and functions it requires, 40k servers does not surprise me at all.

Outside of just the brick-and-mortar stores you have Marketing, Retail, eComm, Merchandising, Strategic Sourcing, FP&A, Finance & Accounting, Asset Protection, Corporate Real Estate, Retail Real Estate, Internal Audit, Supply Chain, Transportation, Business Services, Data Science, etc etc. and IT at every level of those. Each one of these components is large enough to be a medium-to-large sized company in its own right.

You’re ignoring general heavy workloads such as observability. How much telemetry do they gather and analyze for tracking and fraud detection. A quick google on Tesco engineering shows that they process 35k qps against couchbase, and 35 terrabytes of telemetry data per day. They track 150k devices in their ecosystem, which, reading between the lines, would produce the telemetry and require observability, state management, anomaly detection, etc. They have hundreds of thousands of employees using these devices for varying purposes. We’re talking quite a bit of compute, which also requires high availability.

I know nothing about Tesco beyond that quick google search, but I’ve been at several companies where I would read online comments claiming we could reduce our workload to a few servers, and I would think of our tens of thousands of fully loaded machines and roll my eyes.

Reply on news.ycombinator.com