Redefining Virtualization in the VMware Acquisition Era [CL89]

Posted on Thursday, Jan 18, 2024 | Series: Chaos Lever
Ned and Chris tackle the complexities of VMware’s post-acquisition era, offering insightful strategies for enterprises and SMBs!

Transcript

Redefining Virtualization in the VMware Acquisition Era

[00:00:00] Ned: I had an epic breakfast of chicken and waffles this morning. So, I’m not saying that my energy level is going to be a little bit low, but it might. I might be tipping on the edge of a food coma.

[00:00:14] Chris: Yeah, that’ll happen. Did you do that on purpose to torpedo the show?

[00:00:19] Ned: Of course. I’m here to sabotage both myself and you.

[00:00:33] Hello, alleged human, and welcome to the Chaos Lever podcast. My name is Ned, and I’m definitely not a robot. I am a real human person who enjoys having wet feet, cold appendages, and dry dermis. Those are things humans like and particularly enjoy about the winter, right? With me is Chris, who is also here. Hi, Chris.

[00:00:55] Chris: I would just like to remind the audience that at no point have I ever accidentally or otherwise fallen into a river, a lake, a creek, a stream, a bramble, a bush…

[00:01:08] Ned: I don’t see why you’re bringing that up. I don’t see how that’s relevant.

[00:01:11] Chris: I have frequently fallen into disarray.

[00:01:13] Ned: [laugh] More of a constant state of being than anything else.

[00:01:18] Chris: Yeah, it kind of references to self-destruction.

[00:01:21] Ned: Oh, a callback. Well done.

[00:01:23]

[00:01:23] Chris: But you know… cheers.

[00:01:25] Ned: I have been enjoying the flooding we’ve experienced recently.

[00:01:28] Chris: [frustrated grunt]

[00:01:30] Ned: Mmm, oh—

[00:01:31] Chris: what was the question?

[00:01:32] Ned: [laugh] Just an observation that I’m really enjoying the flooding that we’ve had recently, and it’s made it super fun to run near creeks or lakes.

[00:01:42]

[00:01:42] Chris: Yeah.

[00:01:43] Ned: Mm-hm.

[00:01:44] Chris: Yeah. You know, there’s always a crazy option of not running near creeks or lakes in flood

[00:01:49] conditions.

[00:01:50] Ned: Have you seen where I live [laugh] ? I’m right next to a creek. Oh, we’re lousy with it. You’d think we lived in… I don’t know. What’s the land of a thousand lakes?

[00:02:02]

[00:02:02] Chris: Italy.

[00:02:03] Ned: That’s the one. That’s what I was thinking. Right next to the Leaning Tower of Pizza.

[00:02:07] Chris: Mmm, now I’m hungry.

[00:02:10] Ned: [laugh] Some people are very angry [laugh] right now. Let’s make them more angrier. Let’s talk about VMware.

[00:02:16] Chris: Perfect.

[00:02:17] Ned: VMware is dead. Long live virtualization.

[00:02:22]

[00:02:22] Chris: And probably actually still VMware. But—

[00:02:24] Ned: Yeah. I mean, I’m… I am being a tad hyperbolic, and I’m, at best, just bolic, if we’re being honest. But there can be no doubt that Broadcom has been making some moves post-VMware acquisition, and some folks are not going to like it. But it’s not like they can just ditch VMware, right? Or can they?

[00:02:48] Chris: Dun dun dun

[00:02:49] dun-dun.

[00:02:49] Ned: No. No, they can’t. No, it’s—

[00:02:52] Chris: oh.

[00:02:52] Ned: Actually pretty hard [laugh] . So, I wanted to take a moment and seriously consider, what would it take to migrate off of VMware and look at it through two different perspectives or lenses. First, let’s think about what it would mean for a small or medium business that only has, like, a handful of ESXi servers and has an axe to grind when it comes to VMware. And the second is an enterprise, so, you know, 10,000 seats, 20,000 seats that’s running hundreds or possibly thousands of virtual machines across multiple clusters. These are slightly different customer types.

[00:03:28]

[00:03:28] Chris: A little bit. A little bit.

[00:03:29] Ned: Yeah. So, just to remind the listeners what Broadcom has been up to—because they’re fun—one thing that they’ve gotten rid of is the perpetual license, which was, previously, you could just buy a license for ESXi in perpetuity, and the only thing that you would pay for is a renewal of software and service, which would mean you’d get software updates, and you’d get support from VMware. That license is no longer an option, off the table, and when the current S&S contracts expire for existing customers, they will have no choice but to sign up for an annual license, essentially. That’s cool.

[00:04:13]

[00:04:13] Chris: Best way to think about it is, you used to be able to buy VMware, and just have

[00:04:21] it—

[00:04:21] Ned: And now you rent it.

[00:04:23] Chris: That was what was known as ‘owning what you paid for.’

[00:04:26] Ned: [laugh] In fairness, if you read the licensing terms, you never truly owned it. But, like—

[00:04:32]

[00:04:32] Chris: Well, does anybody truly own anything?

[00:04:36] Ned: What even are hands?

[00:04:38]

[00:04:38] Chris: Can you see the colors of the wind?

[00:04:41] Ned: Can you catch a rainbow? The answer to all these things and more are—no. So, they did that. Was fun, getting rid of the perpetual license. They’ve also completely disemboweled the Partner Program, so all the resellers for VMware who were partners, they are going to be no longer partners in about two weeks [laugh] . The partner program is being completely dissolved, and there will be an invite-only Broadcom partnership. So, ‘don’t call us, we’ll call you,’ is basically what Broadcom said. In addition, they’re getting rid of the Cloud Services Partner Program, and I think that’s just gone.

[00:05:19]

[00:05:19] Chris: Yeah, I think they just terminated the concept altogether.

[00:05:23] Ned: Mm-hm, yep. And they’ve offloaded some of the other product lines or announced their intention to offload some of the other product lines. So, the security products that VMware has, like Carbon Black, those are going to be jettisoned off into their own thing, and so will Horizon, which is their VDI slash client management solution. That’s all going to be sold off to somebody else and no longer under the VMware umbrella. Which might actually be good for those products, to have somebody else who actually cares.

[00:05:54]

[00:05:54] Chris: Right. I mean, that’s one of the things that, accidentally or otherwise, is going to probably be beneficial. VMware just has too many SKUs. There’s too many products, there’s too many alternatives, it’s very confusing at this point. Like, if you look at the full menu [laugh] of license options and products and, you know, programs that you can get a part of, it’s absurd how many different things you can do.

[00:06:18] Ned: Yes. So, on the bright side, Broadcom is going to narrow down those SKUs to a lot less, and eliminate a lot of the custom SKUs that were out there. It’s going to be, you can have any color as long as it’s black or white. So, with that in mind, Broadcom is not really interested in SMBs. If I’m going to be blunt, and I could speak to small and medium businesses as if there are people, Broadcom doesn’t give a shit about you.

[00:06:45] Chris: Well, I mean, don’t soft pedal it. Tell it like

[00:06:47] it is.

[00:06:50] Ned: [laugh] SMBs are the most work for the least amount of money, as anyone in consulting can tell you. Your smallest customers tend to be the most needy and also the most miserly. They’re already running on a shoestring budget, they are stretched too thin across too many IT domains, and they have no economies of scale to speak of. The SMB customer is a consultant’s least profitable customer, but they’re necessary for two important reasons. One, I mean, they still are profitable. They add to the bottom line. And two, someday, they might be a bigger customer. Some businesses grow or—and this is more likely—the employees of that SMD will go on to work at bigger shops, and if they like you, they’ll bring you with them.

[00:07:41] Chris: I mean, I think you’re also missing point three.

[00:07:44] Ned: Mmm?

[00:07:44]

[00:07:44] Chris: There’s a lot of them.

[00:07:47] Ned: There are a lot. That’s a really good point. You have to stretch out your sales team to talk to all these different SMBs or rely on a partner to do it because you don’t want to expand your sales team, and as we already discussed, that partner program is very much diminished.

[00:08:04]

[00:08:04] Chris: Good point.

[00:08:06] Ned: I think the second situation of someone learning about a—well, learning about a product or working with a great consulting group and then going to another company and kind of bringing them with them, that’s probably the more important factor. And I think that was a big part of VMware’s overall success. They catered to the SMB as well as the enterprise early on, and they got sysadmins like me hooked on their product. I worked at, like, a 250-person company. I got hooked on VMware. When I moved to bigger organizations, I was like, “This product rocks. If we’re not already using it, we should use it, and if we are using it, we should use more of it.”

[00:08:47] Chris: Yeah, I have a feeling that’s pretty common—especially for the time period that we’re talking about, a pretty common runway.

[00:08:54] Ned: Yeah. So, while working with small and medium businesses is a huge pain in the ass, it’s also an investment in the future as those folks move on to larger organizations that have substantially deeper pockets. So, end of that rant.

[00:09:11] What would it actually mean for one of these SMBs to migrate off of VMware? And the key point is that most SMBs don’t actually have that many virtual machines, relatively speaking: on the order of tens or maybe a hundred. So, the actual migration isn’t that tough. They also don’t have a ton of time or free resources. Most IT departments at an SMB are already running pretty lean, so any replacement solution has to be really easy to migrate to and not have to steep of a learning curve. It has to be very friendly.

[00:09:49] So.

[00:09:49]

[00:09:49] Chris: Right. This is in, especially, situations where you have the IT guy, slash gardener, slash front desk receptionist.

[00:09:56] Ned: Don’t forget about the vending machines.

[00:09:57] [laugh] .

[00:09:58] That small company I worked for that was 250 people, we also were responsible for managing the vending machines. Because somebody had to.

[00:10:06]

[00:10:06] Chris: Hey, that’s a major profit center right there.

[00:10:09] Ned: You’re not kidding. So, taking all that into account, I’d recommend that most SMBs just move their shit to the cloud and adopt as much Software as a Service as possible. Adopting cloud is going to reduce your administrative overhead, you’re no longer managing hardware or a data center, if you have a tiny little data center and a closet, and I know you do. The cloud has been designed as very simple to operate, and there’s migration tools to move your virtual machines and applications. I used Azure Site Recovery Manager to migrate virtual machines to Azure back in 2017. That was super simple to set up and run. I have to imagine that the migration tools have only gotten easier since then.

[00:10:55]

[00:10:55] Chris: Well, first of all, it’s just called Azure Site Recovery. Site Recovery Manager is a VMware product.

[00:11:01] Ned: Maybe—okay, fine.

[00:11:02] Stupid.

[00:11:03] Ohhh [laugh] .

[00:11:05] Chris: But yeah, I mean, you’re absolutely right. And all that stuff comes from pure virtualization technology anyway, from Physical to Virtual, P2V Migrations, or upgrade types of things from one ESXi host to another, V2V migrations. All those tools are things that people are comfortable with, and using something like Azure Site Recovery, or even the easier, more orchestrated version, which is just—at least the last time I looked—was called Azure Migrate.

[00:11:32] Ned: [laugh] .

[00:11:32] Chris: Hey, every once in a while they get the

[00:11:34] naming right.

[00:11:35] Ned: I’d say that Microsoft always names their products exactly what they do.

[00:11:40] Chris: Don’t make me say the E word.

[00:11:42] Ned: Don’t. That’s not nice. You’re not nice.

[00:11:44] Chris: I’ll do it.

[00:11:46] I won’t… exit.

[00:11:48] Ned: Go away [laugh] .

[00:11:50] Chris: Anyway, the point is, I agree.

[00:11:52] Ned: Excellent. Yeah, when you and I were doing consulting, the SMB migration to the cloud was already well underway anyhow. The only things left in most SMB offices are, like, a print server and a file server and the file server is probably gone now, too because people are using Dropbox, or Box, or OneDrive, or Teams, or whatever the hell they’re putting their files in these days. So, you got a print server. That’s it. You can run that on, like, an access point these days, let’s be honest.

[00:12:22] So, if you’re an SMB who has not done that migration yet, and you’re looking to replace your vSphere cluster with another solution, I’d say that, based off of what I saw in the community, a solution like Proxmox, or Xcp-ng would be your best bets for something that is almost free. If you don’t mind paying for a solution, then Nutanix or Scale Computing are very interested in moving you off of VMware, and they’ll probably provide the professional services to do so, which you’re going to need. But seriously, probably just move to the cloud.

[00:12:58] Chris: It really should be the first idea. At the very least, strongly consider it.

[00:13:04] Ned: Indeed. So, that’s SMBs. And Broadcom has no interest in maintaining their SMB relationships. They’ve made that pretty clear. However, enterprises are a different matter entirely. Enterprises are cash cows that move very slow and have large budgets. And there’s less of them. So, for a rent seeker like Broadcom, they are the ideal customer. Why? Well, you have thousands of VMware VMs, and migrating to another platform is not a project to be undertaken lightly, especially since you’re probably not just using VMware, vSphere, and ESXi; you’re probably also using all the other features and services in the VMware portfolio, as you alluded to, Chris, is gigantic. And you have all kinds of other applications that have VMware-specific integrations. Let’s cover a few of them, shall we?

[00:14:02] Chris: Actually, let’s cover all of them.

[00:14:04] Ned: Yes. Welcome to hour one.

[00:14:06] Chris: This show will be 900 minutes long.

[00:14:08] Ned: [laugh] Hour one of a nine-hour podcast [laugh] . You may not leave, you may not pause.

[00:14:13] Chris: There will be no breaks.

[00:14:15] Ned: And you can’t listen to it at two-speed. So, you’ve got stuff like basic compute virtualization. So, that’s your vSphere and your ESXis. But you also have storage virtualization. You may be using vSAN, so you’d have to get off of that. Or you could be using storage arrays that have VMware-specific plugins that you’ve paid for. Those are not going to work with whatever solution you move to.

[00:14:40] Then there’s network virtualization. If you’ve implemented NSX across all your clusters, that’s pretty useful, especially if you’re trying to do micro-segmentation and distributed firewalls, but NSX is not going to work on whatever solution you decide to move to. Backup and recovery. Now, you might not be using VMware’s Tools. In fact, I hope you aren’t, but you’re probably using a backup product that has deep VMware integration, like CommVault, Veeam, or Rubrik. The list goes on well beyond that. Don’t feel like you’re being slighted if I didn’t mention you.

[00:15:14] Then there’s monitoring. You could be using vRealize Operations, which I’m sorry—and I think it’s called something else now. ARIA? Didn’t they rename all their vRealize stuff, like, ARIA? Whatever. Anyway, you could be using vROps, you could be using CloudHealth, or you could be using a third-party tool that keeps track of your host VMs, storage, and networking. You would need that monitoring solution to support your new solution, and reconfigure all your dashboards and everything to point at that new solution and give you useful insights.

[00:15:46] There’s also disaster recovery. Now, God help you if you’re heavily invested in, like, Site Recovery Manager, or you’re using Zerto—talk about deep integration there—but can you imagine overhauling your entire DR tool set while also trying to migrate all your virtual machines to a new platform? That sounds great.

[00:16:07] Speaking of tooling, you also have all your automation tooling. Maybe you’re deeply invested in vRealize Automation. Again, I’m sorry, if you are, but even if you’re not, you probably cobbled together all kinds of VMware-specific automation with power CLI scripts, or baked operations into a ticketing system like ServiceNow, and all those integrations and automations will need to change.

[00:16:32] But we’re not done. There’s VDI. Oh, what’s that? You’re using XenApp or XenDesktop or Horizon? So, now you have to migrate all of these virtual desktops and get feature parity on some other hypervisor and VDI platform. Yeah, that’s going to be fun.

[00:16:52] And then lastly, you have operations and process. Now, this isn’t even a tooling conversation as much as it is a training and updating existing processes for the new platform. Your ops team has over a decade of experience with VMware. Are you really going to disrupt their flow to move them to a new platform? You already brought in cloud at some point, and they’re still pretty pissed off about that.

[00:17:17] Chris: So, what you’re saying is, this is a pretty easy migration.

[00:17:20] Ned: Yeah, you just do it overnight. It’s no problem. Zero downtime. When you take all these factors into account, you can kind of see how Broadcom has enterprises over a barrel. Migrating to a different platform is going to be incredibly expensive, and time-consuming, and ultimately provide very little in short-term business value.

[00:17:44] Imagine trying to sell that up the chain in this economic climate, versus just accepting the end of perpetual licensing and allocating, like, an extra 10% for VMware slash Broadcom in your budget? If I’m, like, a CIO or a CTO, and I’m working on the future-looking technology strategy, I am way more focused on, like, AI than trying to replace the virtualization platform because Broadcom wants to fleece me for, like, an extra 100 grand.

[00:18:15]

[00:18:15] Chris: Yeah…

[00:18:16] I mean, you’re right because the problem here is much larger than just the bill going up by X amount percent, for no tangible gain. All that stuff has to change, it has to change in a particular order, if it doesn’t change successfully, you are now in a crisis of varying different sizes, depending on what failed. Certain tools are probably just immediately disqualifying.

[00:18:46] Ned: Yep.

[00:18:47] Chris: If your migration includes ‘migrate off of NSX,’ that’s going to take a minute.

[00:18:54] Ned: [laugh] Yeah.

[00:18:54] Chris: You remember the four years it took you to get on to NSX?

[00:18:58] Ned: It ain’t easier getting off!

[00:19:00] Chris: No, it is not.

[00:19:02] Ned: Mmm. So, from a pure business perspective, and that’s really where we need to look at this—like, from a philosophical perspective, you might be very angry that you’re now being asked to pay more money for less innovation. I get that. You may be affronted morally, but we live in a capitalistic society, we have to think about the business, and the right AI integration could make or break a company. Moving to a different virtualization platform is a boondoggle that could end my personal career as well as sink the company. So, which one am I going to focus on?

[00:19:39] Chris: Right.

[00:19:40] Yeah, I mean, it’s the standard conversation, right? It’s risk versus reward. The reward is getting away from being ripped off, more or less, by a company that’s going to pull as much money out of you as humanly possible, and stop really increasing the quality of the product, reasonably because as we talked about many, many times, ESXi is complete.

[00:20:08] Ned: Yes.

[00:20:09] Chris: We’re done.

[00:20:10] Ned: Yep.

[00:20:11] Chris: It’s good.

[00:20:12] Ned: Aside from, like, security patches.

[00:20:14] Chris: Right? Yeah, yeah, yeah.

[00:20:15] Ned: We do not have to add anything, nor should we.

[00:20:18] Chris: And that’s the important part, actually. Nor should we.

[00:20:20] Ned: Yeah.

[00:20:21] Chris: Anyway.

[00:20:22] Ned: So, I guess the answer is, enterprises, sorry, you’re fucked. I don’t want to sugarcoat it for you, but that’s kind of where we’re at.

[00:20:32] Chris: At the very least, it’s going to take a while—

[00:20:35] Ned: Yes

[00:20:35] Chris: —and this is a program or project that you’re going to have to approach very carefully and stepwise.

[00:20:42] Ned: And have justifications for doing it beyond, “We don’t like Broadcom.” Unless you’re the CEO, and then you can be like, “Well, I just don’t like that company, and we’re not using it.”

[00:20:51] [laugh] .

[00:20:52] Oh, there’s a story about eBay and Solaris that we’re not going to get into. Anyway. So. While that was some serious worst case scenario stuff, I don’t think things are actually that bleak. Taking on a massive migration effort across 100 or 1000 VMware clusters is probably going to be a quagmire from which your company and your career never fully recovers. Just look at the long trail of failed ERP migrations that turned into case studies for your MBA. You can be a case study. But you don’t have to eat the whole VM migration elephant in one fell swoop.

[00:21:28] This is more of a death to VMware by a thousand cuts. And I think some of those cuts will fall under the categories of app modernization, SaaS adoption, and replatforming. So, if you’re trying to prioritize things, app modernization is probably the best answer. Migrate your apps off of VMware, not with new virtual machines on a different platform, but move it to a cloud-native architecture, by which I do mean more than just shoving everything into containers, throwing it on Kubernetes, and calling it a day. We’re actually talking about adopting Platform as a Service and Kubernetes for your applications, rewriting them as needed, and doing this one application team at a time.

[00:22:16] Is that a massive undertaking? Yes. But unlike simply shuffling the application into a new virtualization platform—which is basically what we did with P2V—modernizing an application can actually provide real business value in terms of cost, performance, and functionality. So, if you’re going to prioritize moving off of VMware, I would start with application modernization.

[00:22:41] Chris: Right.

[00:22:41] Especially if your applications haven’t been updated or modernized in a while, this is going to provide benefits outside of just getting away from VMware.

[00:22:51] Ned: Absolutely.

[00:22:52] Chris: Modern applications are built—even the ones that are commercial off the shelf—they are built to be resilient, and redundant, and completely indifferent to geography.

[00:23:05] Ned: Exactly.

[00:23:06] Chris: Which is a huge benefit for all the other things we talked about up top: disaster recovery, high availability, the ability to migrate without creating huge downtime windows. A lot of monolithic software that runs on-prem was never intended for that, so redoing it, whether you literally rewrite it or you just replace it with something better, you get business operating benefits in addition to not being reliant on your Infrastructure as a Service on-prem solution, aka VMware.

[00:23:35] Ned: Right. And going on that thought of, there are a ton of commercial off-the-shelf applications that are now available to SaaS. That’s probably what you should focus on as well. Chances are, you probably already migrated all the low-hanging fruit onto a SaaS platform: your email, your video conferencing, file sharing, et cetera, all that is probably already on Microsoft 365, Zoom, Box or, you know, something similar G Suite. I don’t know. I’m not here to judge. Maybe a little.

[00:24:07] So, now it’s time to evaluate the other applications you have and see if they’re also a good fit for a SaaS product. If something that you’re running internally, is a commercial off-the-shelf piece of software, and it’s running on VMware today, then 99% of the time there is a SaaS version of that same software that you can just migrate to. And unless you’re in a regulated industry that has to use hosted services—and those are somewhat rare—all your COTS applications should be farmed out to SaaS, and that probably includes your VDI farm. Get rid of that crap too.

[00:24:44] Chris: That’s a controversial take.

[00:24:46] Ned: It is a controversial take, but as someone who’s used some of the cloud-based VDI solutions, they make your life so much easier. Boot storms are a thing of the past, my friend.

[00:24:58] Chris: That’s the only time I feel alive.

[00:25:02] Ned: I’m, I’m sorry? [laugh] . But ultimately, there are going to be some virtual machines that cannot or will not be modernized or migrated to SaaS, and those are the ones that after you’ve taken care of those first two steps, you can now move to another virtualization platform. This should definitely be at the bottom of your list. Get all your other applications off of VMware first. Suddenly, you have a lot less VMware to deal with.

[00:25:27] So, the last part I want to talk about is the alternatives to the world of VMware. There are a few, but I’d also recommend checking out a Reddit discussion I found that talks about each of these with people who have actually run them at scale, and they have opinions.

[00:25:45] Chris: On Reddit?

[00:25:46] Ned: I know. Shocking. Are they correct opinions? [pause] Eh [laugh] . But they are opinions. So, let’s start with Nutanix. This is the obvious alternative. Nutanix has been pushing their AHV—is that Acropolis? Is that what they call their hypervisor?

[00:26:06] Chris: Yeah, I think so.

[00:26:07] Ned: They’ve been pushing their Acropolis hypervisor for years, and touting their superiority in terms of software-defined storage. I’ve always found their sales and marketing teams to be a bit distasteful, but the technology seems good, and they are very focused on getting you off VMware, so they might help.

[00:26:24] Chris: Yeah, in terms of institutional or industrial-sized solutions, this is probably the solution.

[00:26:29] Ned: Yes.

[00:26:30] I’ll mention another industrial-sized solution, is—that you have to pay for, and that would be Hyper-V. And yes, I am serious. Stop calling me Ballmer. Hyper-V is the underpinning for Microsoft’s Azure virtualization, and they use Storage Spaces and Storage Spaces Direct for their storage, and they have a software-defined networking stack that is actually pretty robust, and it integrates with Sonic, which is an open-source, software-defined, like, wide-area networking. And I guess it does some orchestration one—it’s a lot. What I’m saying is, if you don’t mind paying for Windows licenses, or you already are—which you probably are—Hyper-V is actually a pretty solid option. I can’t believe I’m saying that out loud, but I am.

[00:27:16] Chris: Well, they haven’t done the best job of really selling it in that way, I think was part of the problem.

[00:27:22] Ned: Yeah.

[00:27:22] Chris: People are always like, “Hyper-V? Oh, all right, I’ll use that to run a virtual machine on my laptop.” But it can do a lot more than just that.

[00:27:32] Ned: Well, that’s what I’m saying. Like, they’ve never done a great job at marketing Hyper-V, especially in the US, for some reason. I ran across a bunch of people over in the EU that had been running Hyper-V for years since, like, Server 2008, and they’re very happy with it. And I’m like, shouldn’t those people come over and maybe tell some folks in the US that this is the thing? But anyway.

[00:27:53] [laugh]

[00:27:54] Chris: give one of these guys a microphone at Ignite.

[00:27:57] Ned: For real [laugh] . So anyway, those are, like, the paid and enterprise-familiar options. Some open source options, we have Proxmox, which is fun to say. This was the virtualization platform mentioned the most often in the Reddit posts that I looked through. It’s based on KVM and LXC.

[00:28:20] It’s open-source and free to use, and it has no enterprise features locked behind a paid license. You get everything upfront. If you’re going to use it, you can pay for support if you want it, which if you’re an enterprise, you probably do. If you want software-defined storage, it uses Ceph, so you have to be comfortable with Ceph, or bring your own SAN array or NFS device. But Ceph works fine, from what I hear.

[00:28:50] In terms of networking, it uses open vSwitch, and some more recent implementations offer software-defined networking that includes things like VXLAN, BGP, all the, kind of, fun things that you might want in a software-defined networking package. Which means you could—since you can’t just migrate NSX over, but you can get some of the same features.

[00:29:13] Chris: Right. You would probably have to build side by side—

[00:29:16] Ned: Right, absolutely.

[00:29:17] Chris: —and have to do a cutover, not an actual migration live.

[00:29:20] Ned: Yeah.

[00:29:21] Chris: And networking people are going to love this idea, so bring it up… first thing in the

[00:29:26] morning.

[00:29:27] Ned: [laugh] Before they’ve had their coffee. So, based on some folks who are running this at scale, it appears to be a pretty solid option to VMware, but it will require a bit more care and feeding. So, it’s free like a puppy, as it were. Also, the UI is pretty hideous and very orange.

[00:29:47] Chris: What do you have against orange?

[00:29:49] Ned: Nothing. I just think there’s a time and place [laugh] . That time is less. The next one is called Xcp-ng, which is brought to you by people who hate naming and also vowels. I don’t know if there’s a way to pronounce this that’s not just saying the letters in order. [makes a valiant try: exapng]

[00:30:10] [another valiant try: chip-ngh] .

[00:30:11] It’s XenServer. That’s all it is. Maybe the NG is for, like, ‘next generation?’

[00:30:15] Chris: why didn’t you just say that then?

[00:30:18] Ned: So, it’s part of the Xen Project that’s part of the Linux Foundation, and so I think XCP is, like, XenServer Control Platform or something like that. And then N-G is for next generation, but just, like, no. Ever—all of this is bad. Like, XenServer is a pretty good name. Couldn’t you just use something like that?

[00:30:38] Chris: Guess not.

[00:30:40] Ned: [laugh] Naming is hard. It’s fine. So, if you remember XenServer, it was kind of the third-party vote for virtualization. If you ever had to take a Xen app or Xen desktop course, you probably were forced to learn about XenServer. Now, it never really went away. AWS actually used it for years before moving to their own KVM hypervisor, and Xen, like I said, is now part of the Linux Foundation.

[00:31:08] And in terms of this Xcp-ng, there are no enterprise feature gates that are behind paid licenses, just like Proxmox; you get the whole functionality right out of the gate. There’s paid support if you want it. This seems to be a close favorite second behind Proxmox, and I suspect the name is [laugh] why it’s in second place instead of first.

[00:31:30] Chris: It’s really—yeah, it’s a rough one.

[00:31:33] Ned: Yeah.

[00:31:34] Chris: It looks like somebody sneezed and slapped the keyboard by

[00:31:36] accident.

[00:31:37] Ned: Kind of. Or just banged their head [laugh] . They got frustrated and banged their head, and they’re like, “Oh, that’s good name. We’ll go with that.” Next one is called oVirt. O-V-I-R-T. This is the free and open-source version of Red Hat virtualization. Just like Proxmox is KVM-based, and since it’s related to Red Hat, it uses other Red Hat tools like Ansible and Gluster for the automation and storage, respectively.

[00:32:04] For networking, you can use the built-in oVirt networking, or you can use OpenStack’s Neutron Networking if you need more advanced features. A lot of the folks in the comments seem to think that Red Hat is less interested in maintaining oVirt, and I can’t speak to that. Maybe that’s true. But a few heavy hitters also called out that they’re running oVirt across their entire data center, and it works just fine. So, if you’d like Red Hat projects, it might be worth giving it a look.

[00:32:33] Chris: Right.

[00:32:34] Ned: Not even an honorable mention, just a funny mention is OpenStack because basically, no one mentioned it as a viable alternative. I mean, Red Hat really likes OpenStack and apparently telcos love it, but as I understand it, it’s a pain to manage, and it’s overkill for most people, and if you don’t have a dedicated engineering team, or a Red Hat support contract, it’s probably best to look at one of the other solutions that I mentioned.

[00:33:00] Chris: Right. It’s interesting, the order that you’ve put these in, particularly with OpenStack at the bottom. One of the biggest things about VMware—in particular ESXi and vSphere—next, next, next, finish.

[00:33:15] Ned: Yes.

[00:33:16] Chris: You’re configured.

[00:33:17] Ned: Mm-hm.

[00:33:18] Chris: It really, it just works, the user interface is very solid, even though it’s weird. We had a little—you know, there was a rough period there when we moved to HTML5 that a lot of people don’t like to talk about. But all these other ones, the technology underneath of it, it all does the same thing. It’s a matter of how you get there, and how do you keep your eyes on it, and how do you make changes if and when you need to? So, a lot of these, like you said, it’s free like a puppy. You’re just going to have to feed the puppy, you’re going to have to walk the puppy, sometimes you’re going to have to clean up after the puppy.

[00:33:53] Ned: Right.

[00:33:53] Chris: Whereas OpenStack is more like, ‘free like a tornado.’

[00:33:57] Ned: [laugh] I was wondering what metaphor you’d go for. That’s perfect [laugh] . Oh, God. Yeah, no, that’s accurate. I have used… not all of these. I’ve installed Proxmox, and Hyper-V, and Nutanix, and OpenStack at some point, and I used XenServer once, and I got to say, like, in terms of user interface, XenServer actually has a very friendly and pretty straightforward user interface.

[00:34:26] And based off my limited experience with Proxmox, it also has a pretty nice UI that’s very friendly to the new user. So, this is not the dark, awful time of ten years ago where it was either command line or get the fuck out. All of these platforms have matured and added more friendly features for new users who are expecting more of a VMware-like experience.

[00:34:50] Chris: So, if you had to pick?

[00:34:52] Ned: Right now, I would probably pick Proxmox if you’re, like, an open-source shop who cares about that, and Nutanix if you aren’t.

[00:35:02] Chris: Simple enough.

[00:35:03] Ned: Yeah, that’s the shortlist of virtualization platforms. By no means is it meant to be exhaustive. There are some platforms I saw folks talking about and actually being used at an enterprise scale. That’s the ones we just mentioned. There were also a few people who have built their own homegrown solution from open-source parts. That’s a choice. You do get infinite customizability and flexibility, but you’re now on the hook for your own support, and you have to train every person you hire to use your weird system. There’s trade-offs is what I’m saying.

[00:35:39] So, [sigh] enterprises are deeply invested in VMware. I don’t see that changing in the immediate future, but I think there is going to be a slow move away from VMware over the next five years.

[00:35:52] Chris: Yeah, that makes sense. That’s the kind of time where, like, I would guess that a lot of companies will pay the Broadcom tax—

[00:36:03] Ned: Mm-hm.

[00:36:03] Chris: —once. Whatever contract they sign, whether it’s a three-year contract or a one-year whatever, probably the last one.

[00:36:12] Ned: If not the last one, the second-to-last one, and then they will start firing up their efforts to get off of VMware.

[00:36:19] Chris: Right.

[00:36:20] Ned: And you’re going to have a new—a fresh crop of people who are coming in and up from the small businesses that either have never touched virtualization because everything’s cloud-native, or they already went through this migration process, and they’re ready to help you.

[00:36:36] Chris: Speaking of getting off of VMware, how about you get off the microphone, eh?

[00:36:42] Ned: Ah-ah. Yeah, thanks [laugh] . Great segue. Thanks, dude. Well, thanks for listening, or something. I guess you found it worthwhile enough if you made it all the way to the end, so congratulations to you, friend. You accomplished something today. Now, you can sit down on the couch and brainstorm better names for Xcp-ng. I am sure that you can do it. You can send those to Chaos Lever. Just go to our website, chaoslever.com. We have a contact form. You can tell us your cool new name, or something you want us to cover on the show. That would be fine. You can also check out our LinkedIn page. Just search Chaos Lever on LinkedIn, or like I said, you can go to the website. We’ll be back next week to see what fresh hell is upon us. Ta-ta for now.

[00:37:36] Chris: I think this should just completely go in the opposite direction. Everybody’s trying to be all serious with these names. What if Xcp-ng renamed it, like, CatPoop?

[00:37:44] Ned: [laugh] You mean like OpenTofu?

[00:37:47] Chris: Yeah, but, like, after the fact.

[00:37:50] Ned: [laugh] .

Show Notes

Redefining Virtualization in the VMware Acquisition Era

Episode: 89 Published: 1/18/2024

VMware’s virtualization Voyage

Ned and Chris are at it again in the latest Chaos Lever episode, serving up their unique blend of tech expertise and uproarious humor! This time around, they’re tackling the big shake-up in the virtualization world – the aftermath of Broadcom’s acquisition of VMware. The duo dissect how these changes impact different business scales, from SMBs to large enterprises, providing actionable insights and practical strategies. If you’re grappling with decisions about virtualization, cloud migration, or seeking alternatives to VMware, this episode is your goldmine.

Intro and outro music by James Bellavance copyright 2022

Hosts

Chris Hayner

Chris Hayner (He/Him)

Our story starts with a young Chris growing up in the agrarian community of Central New Jersey. Son of an eccentric sheep herder, Chris’ early life was that of toil and misery. When he wasn’t pressing cheese for his father’s failing upscale Fromage emporium, he languished on a meager diet of Dinty Moore and boiled socks. His teenage years introduced new wrinkles in an already beleaguered existence with the arrival of an Atari 2600. While at first it seemed a blessed distraction from milking ornery sheep, Chris fell victim to an obsession with achieving the perfect Pitfall game. Hours spent in the grips of Indiana Jones-esque adventure warped poor Chris’ mind and brought him to the maw of madness. It was at that moment he met our hero, Ned Bellavance, who shepherded him along a path of freedom out of his feverish, vine-filled hellscape. To this day Chris is haunted by visions of alligator jaws snapping shut, but with the help of Ned, he freed himself from the confines of Atari obsession to become a somewhat productive member of society. You can find Chris at coin operated laundromats, lecturing ironing boards for being itinerant. And as the cohost on the Chaos Lever podcast.

Ned Bellavance

Ned Bellavance (He/Him)

Ned is an industry veteran with piercing blue eyes, an indomitable spirit, and the thick hair of someone half his age. He is the founder and sole employee of the ludicrously successful Ned in the Cloud LLC, which has rocked the tech world with its meteoric rise in power and prestige. You can find Ned and his company at the most lavish and exclusive tech events, or at least in theory you could, since you wouldn’t actually be allowed into such hallowed circles. When Ned isn’t sailing on his 500 ft. yacht with Sir Richard Branson or volunteering at a local youth steeplechase charity, you can find him doing charity work of another kind, cohosting the Chaos Lever podcast with Chris Hayner. Really, he’s doing Chris a huge favor by even showing up. You should feel grateful Chris. Oaths of fealty, acts of contrition, and tokens of appreciation may be sent via carrier pigeon to his palatial estate on the Isle of Man.