Author: Scott Kinka

The Bridge Podcast - AJ Kuftic Expedient LogoOn this episode of The Bridge, I’m joined by AJ Kuftic, Product Manager at Expedient. We’re talking about the future of VMware and so much more.

Expedient is a data center and cloud services provider offering infrastructure as a service (IaaS) with local operations in Pittsburgh, PA, Baltimore, MD, Boston, MA, Cleveland, OH, Columbus, OH, Denver, CO, Indianapolis, IN, Memphis, TN, Milwaukee, WI, and Phoenix, AZ. Ranked as one of the Top 3 managed services providers worldwide on Channel Futures’​ MSP 501 list, Expedient’s converged solutions enable clients to focus on strategic business innovation, while the Expedient team handles operation of the information technology needed to support it. Expedient data centers are compliant with the Health Insurance Portability and Accountability Act (HIPAA) as well as the Payment Card Industry Data Security Standard (PCI DSS). Service Organization Control (SOC) reports are published annually for all locations.

During our conversation, we discussed what’s happening with the VMware Broadcom situation and broke down the implications, including the background, impacts on service providers and customers, and the anticipated timeline for adjustment and optimization efforts.

Topics covered in this episode:

  • AJ’s background and career experiences.
  • An overview of the Expedient brand and its evolution of services to meet current market demands.
  • The importance of understanding client pain points.
  • Adaptation strategies for services providers and how the shift from per-gigabyte RAM to per-core licensing facilitates “bring your own licensing” capability.
  • Changes in the private cloud model.
  • The investment in service providers by Broadcom.
  • How VMware’s shift towards bundling products under Broadcom’s ownership contrasts with its previous approach of avoiding shelfware and the impacts of this approach.
  • The impacts of the transition from perpetual to subscription licensing.
  • The general trend towards subscription models.
  • Factors of consideration for alternative platforms.
  • Predictions for the next 12 to 18 months.

Links for this episode:

Productivity AI with Expedient’s Brad Reynolds

The Bridge Podcast - AJ Kuftic Expedient HeadshotABOUT AJ KUFTIC

AJ Kuftic is Product Manager for Expedient. AJ has over 15 years of experience as a customer and partner helping end users build solutions that are sustainable and easy to manage. Having knowledge across various silos of IT infrastructure gives AJ a unique perspective of the pain points and what customers are looking to improve.

When AJ isn’t thinking about the next big thing, he spends his time with his wife and 2 children trying to bake the perfect loaf of bread.

CONTACT AJ

LinkedIn

Web

SUBSCRIBE ON YOUR FAVORITE PLATFORM

          

Full Transcript

Scott Kinka:

Hi, from wherever you are listening to us, and welcome to another episode of the Bridge. I’m your host, Scott Kink. One of my guests today is AJ Kuftic. I said that, right?

AJ Kuftic:

Right.

 Scott Kinka:

AJ Kuftic

AJ Kuftic:

Yes. You got it. 

Scott Kinka:

Okay. I like to not have used that. AJ is a product manager at Expedient who takes a lot of the front-end conversations for that team, and I’m super excited to have him on the show. If you follow US Weekly, you know that we just had Brad Reynolds from Expedia on the show. We had a great episode about AI, and maybe if we get there, we’ll see. But the main reason that we got together and I asked AJ to get on this show is as a product manager there. He’s very involved in what’s going on with the whole VMware Broadcom thing. So we’re going to just get right at it and address that in this episode. But first, we want to learn a little bit about AJ. So we’re going to start right there. Aj, just tell us a little bit about you personally and professionally, how long you’ve been at Expedia, and just give us the download.

AJ Kuftic:

Yeah, so my name is AJ Kuftic. I am a product manager here at Expedient. I have been in the IT industry in some form or fashion for 24 years now, which is a wild thing to think about these grays. They came, honestly. I started my career working in the steel industry, doing IT for process control desktop support, network support, and service support. Moved into pharmaceutical support, then went to a var. So I worked at a partner in the southeast, moved back up here, worked at healthcare IT as a VMware architect, and then landed here at Expedient in December of 2019 with the expectation that I was going to do a lot of this stuff in public. Then, things happened in early 2020, and I didn’t. So it turned into, you explored

Scott Kinka:

Quite a bit about those things that have happened on the show over time. So maybe we’ll dabble a little bit into that. But, so you said steel, you worked in steel, you’re from Pennsylvania?

AJ Kuftic:

I am from Pennsylvania. I grew up in Edinburgh, Pennsylvania. I moved down to Pittsburgh to go to college. My internship was actually with US Steel, so I worked at the steel mills. I wore flame-retardant pants to work. It’s a very fun time when you have to go through all the same safety training as all of the other people in the mill, but your job is to fix computers and not actually make any steel. However, I supported servers, networks, and PCs that supported the entire steel-making process, from coal coming off of barges all the way through to rolled coil. So it was a very, very interesting time. If you ever watch How It’s Made, it’s one of my favorite shows ever. Yeah, it was like going to that every day. So that was really fun and interesting, and it was really bad when things were broken, and you were the person who had to fix it so that things could be made again. But that’s awesome. That was a really fun time.

Scott Kinka:

And looking over your LinkedIn profile, I mean, most of your career has been on the customer side. I mean, you’ve sat in the same seat as the people who you work with on the customer side today. Yeah,

AJ Kuftic:

And I think a lot of that informs how I think about how we deliver our services, what services we want to deliver, and what the problems our clients are facing are. Because I’ve been there, I’ve sat in those chairs, I’ve worked directly with the people who are doing that day in and day out. I was the person doing it day in and day out. I know the pain in a very real way because it was me. So I think that’s something that a lot of folks who have been more in the partner side of the world, the vendor side of the world, sometimes forget, not because it’s a malicious forgetting, but it’s been a while since you’ve done it. And so you kind of lose that. You lose the anxiety, and you lose the feeling of what it’s like to sit on that chair and have to deal with tickets and have to deal with upgrades and patches and projects that you’re only really ever focused on that one thing that you’re doing with that one customer.

AJ Kuftic:

It’s also super interesting to be on the vendor side and the partner side to be able to go and see all of the different places and the different problems that everybody’s facing. So I kind of got to blend both of those sides. I worked with a partner traveling through the Carolinas and Southern Virginia for three years and seeing all of the various different levels of capability as well, which I thought was super interesting. Like, oh, this is a hospital. They probably have really great things. Nope, this one has a carpeted data center, which is a true thing. Or, Hey, this facility that seems really, really small actually has one of the nicest data centers you’ve ever seen. So it was really weird to see these dichotomies and shifts, but at the end of the day, the goal of what we do in technology is to enable our businesses to enable them to actually solve their real problems and actually provide value somewhere. And technology is not really what most companies do for them as a product. It’s not what they said, product. And so when being able to see that and being able to understand where these pieces fit in, I think allows me to have a really unique sort of view on things.

Scott Kinka:

Perfect. So you were Pittsburgh. I’m Philadelphia. You had a little anecdote about Philadelphia. What is it that you know about Philly?

AJ Kuftic:

I know about Philly because my brother lives there.

Scott Kinka:

Okay.  Where in Philly?

AJ Kuftic:

I don’t know exactly where he lives right now. I know. Been all over the place. He’s rented a bunch of places. If my brother is watching this, Hi Casey, I love you. But it’s been different because I’m from western PA. Eastern PA is a different planet. There’s this big ridge of mountains called the Appalachians that runs directly through the middle of Pennsylvania. Suppose you haven’t looked at a map of Pennsylvania for a while. But the mountains really separate the state into two, almost completely separate worlds. Western PA can be, depending on where you are. Southern west, southwest pa. Where I am now is very much Appalachia adjacent, if not directly Appalachia as you go further south. Northwest PA is the Midwest. It is Buffalo. It is Cleveland. It is that ring. But Philly might as well have been New York for us. You always feel like we’re New York’s little brother.

Scott Kinka:

You are New York’s little brother.

AJ Kuftic:

You’re much angrier, and I kind of love it, though. I’ve come to appreciate the sheer chaos of Philadelphia.

Scott Kinka:

We’re going to have a little bit of fun with your brother, though, and I’m not going to share with you how I would respond to this, but has he changed his sports affiliation since moving from Western Pennsylvania to Eastern Pennsylvania?

AJ Kuftic:

Not fully, but yes. 

Scott Kinka:

See, I would’ve thought it would’ve been the other way, and I would’ve fully. The only reason you’re allowed to be in Philadelphia and be a fan of another team is if you moved from there, right? Yes. Or that’s a complete, and I can completely respect that. It’s the people that grew up here who were like, well, I was a Dallas Cowboys fan in the eighties. They were good. You lose, you’re out. See you. Get out of here.

AJ Kuftic:

Get out of here. Get out of here. You’re about to get punched in the face. I got it. No, he grew up; we grew up in Western PA. My dad’s family’s from Pittsburgh, so we were all Steelers fans. Really great. So he moves over there. The Steelers and Eagles don’t play that often, so it’s not that big of a deal, like the Battle of PA Parade.

Scott Kinka:

I would totally go have a beer with your brother and watch a Pittsburgh game and root for him if it was against somebody else because he’s worn in BR, and he gets to come over here, and that’s totally cool.

AJ Kuftic:

But he cheers for the Eagles, too.

Scott Kinka:

Listen, you can have an FFC team, and everybody gets to have a home team and a team from the other conference. I’m good with that, right?

AJ Kuftic:

Gilberts, by the way.

Scott Kinka:

Yeah. Thank you. Hopping back to work, is there a specific phrase around the Steelers? I’ll give it back to you.

AJ Kuftic:

I mean you, here we go. Just here we go. Steelers. There we go.

Scott Kinka:

All well, here we go. Steelers there. Alright, so jumping into just going back to work for a minute. We’ve said Expedient. We’re going to put the episode with Brad in the liner notes here as well for anybody who’s listening to it. But can you just give us the, for those who don’t know the brand, tell us a little bit more?

 

Navigating the VMware Landscape

AJ Kuftic:

Expedient is a full-stack cloud service provider. There it is. Got it all the way out. We basically do infrastructure as a service, so VMware as a service, backups as a service, dr as a service. We started 20 ish years ago doing co-location. We actually started as an ISP that moved into co-location, and then this thing called Virtual co-location came around, which was VMware as a service. And then, hey, people wanted backups. So, we started doing backups. We’ve done web hosting long, long, long ago. But all of those things led to where we are today, which is being an infrastructure as a service provider and, in some cases, a platform as a service provider for things like Kubernetes. So we run this really wide gambit of services that meet the needs of a lot of organizations, especially when it comes to IT infrastructure.

Scott Kinka:

Gotcha. And be completely transparent with our audience. The reason why we have AJ on here, I mean, AJ had mentioned earlier, spent a lot of his career as a VMware architect and landed here at an infrastructure as a service company that has a lot of services, not exclusively, but a lot of services based around VMware and as a product manager, has had to deal with some interesting recent news and we had a lot of our users kind of hit us up online and ask us questions. So this is one of probably a couple of episodes that we’re going to do from a couple of different perspectives talking about the Broadcom and VMware situation let’s say. But for those who aren’t familiar with what’s going on, can you just give us a little bit of historical background on what all the fuss is about?

AJ Kuftic:

Sure. So VMware, if you are in the IT space and you haven’t heard of it, got acquired by Broadcom. VMware has been acquired before. This is not their first time getting acquired. They were acquired by EMC in 2009. Then, EMC was acquired by Dell. And so Dell effectively ran VMware for a while, and then Dell sold off VMware to Broadcom. In the past acquisitions, VMware’s generally been left alone almost hilariously because when EMC owned VMware, it was incredible how many times EMC would come out with some sort of new feature and not support it. VMware or VMware would come out with a feature, and EMC wouldn’t support it. And it’s like,

Scott Kinka:

Oh, I know. Having run a service without it myself with both EMC and VMware products in it, I get you. 

AJ Kuftic:

Yeah, what are we doing here? Right? You guys have. When VMware rolled out virtual volumes, EMC didn’t support it for a solid year or two, which was very funny. Dell, on the other hand, when they owned VMware, tried very hard to integrate VMware into Dell as much as they could. This is where VxRail comes from. So originally, it was Project Marvin, which was the original title of it, but they were taking VMware and Vsan and putting it onto a Dell hardware box that they could then manage together as one unit versus running Visa and yourself on Ready notes. So they had these different things where they were trying to actually integrate it, which was meet, but then they get sold to Broadcom. And Broadcom has made some acquisitions in the past of large companies. You might’ve heard of Symantec and ca. And what Broadcom had done over the last when they announced the change, were they specifically want to raise EBITDA off of VMware. VMware had about a 4 billion EBITDA. Broadcom wanted to take it to eight. And so there’s a number of different ways you can do that. The first one is you start to lower your staff costs. So they have had large layoffs as part of the acquisition, but the other thing is that you need to raise more revenue. You need to have more revenue coming in. And VMware’s product line has been huge. They had about 9,000 SKUs across the entirety of VMware, and Broadcom is cutting it down to about 160. And that is a huge cut. And you would think, how are you raising revenue by cutting the number of SKUs you have? And the answer is they are bundling everything. VMware rolled out Cloud Foundation in 2017, which was effectively NSX vsan, ESX, vCenter, and then a function called SDVC manager that allowed you to deploy clusters with all of the components. If you remember the VCEV blocks, it’s a very similar construct. You have a bill of materials that has this version of vCenter, this version of van, and this version of N sx, and all of them have been IOP tested together, but they added in the piece that VCE never did much to my chagrin in my past, where they actually would automate the actual deployment and upgrades. So all of that comes together to form this one big bundle. And VMware had been struggling to sell Cloud Foundation as an overall bundle because, honestly, it’s expensive and a lot of organizations for years have had perpetual licenses. You bought this Enterprise Plus license in 2011, and you’ve just been paying support for it ever since then. Broadcom said, now we’re getting rid of all that. So they got rid of all the perpetual licensing, and they’re moving everything to subscription licensing entirely to start to recoup the revenue. That, to be fair to Broadcom, had been left on the table by VMware for the last seven-ish years.

Scott Kinka:

Certainly, and just for clarity’s sake, if you are an end user who’s listening to this and has been buying VMware under perpetual licensing, service providers have always paid the way that VMware is asking you to pay today. We had a point system, and points were worth something. Every SKU was worth a certain number of points, and that was negotiated with your distribution or service provider agreement. So the consumption of everything down to individual virtual machines where really, for a long time, it was around the amount of memory that was virtualized.

AJ Kuftic:

It would be still way up until it was still that way up until they’ve just recently, literally in the last few weeks, made the shift for us. So, we are also going to a per-core model. Yeah, Scott’s absolutely correct. The service providers have been using this subscription model since 2007, since the initial rollout of the VMware cloud provider program. It has been really helpful for us to be able to deliver very granular costs to our clients. You were paying for a gig of powered-on RAM and per gig of storage on vsan, so you could get super granular with the amount that you were using, and because you were leveraging a service provider like Expedient, you could only really pay for what you were using versus I have to license the ten hosts in this cluster, even though I’m only really using six and a half of them resource wise.

Scott Kinka:

And just to be clear for a second, it made it. You could granularly sort of build-up that cloud-based bill of materials because the VMware licensing was granular enough that you could embed it into the service prices. Right. So, am I thinking about that the right way? So basically, if I was buying a Giga vsan, it was already loaded with VMware licensing in it.

AJ Kuftic:

Yes. So when you were buying storage from us, that was on vsan. It had the VS a n licensing in it. We also would do things like for Dr. Our clients who were replicating things to do, and they didn’t pay for VMware licensing until the VM got powered on. So, if you’re building a DR structure for yourself, you couldn’t do that. You still had to license the hosts on the other side so that you could fail over. That was a huge cost and was a reason why a lot of organizations turned to service providers for DR because we were able to effectively cut out the VMware licensing costs, and you only had to pay them when you actually turned everything on so you could burst that pricing. It’s a true cloud model. The other part was that we did a lot of it through what are called resource pools in VMware. A lot of people may hear the words resource pool and shutter because, in the early days, resource pools were used as folders, and people didn’t understand how resource pools truly worked. We do. But what we would do is we would give you a set of resources. You would pay us for 500 gigs of RAM and you could spin VMs up and down inside of that pool, and then you just paid for your 500 gigs of RAM overall. Instead of saying, okay, I have three hosts in, each of them have 500 gigs of RAM, so I have failover and the ability to take a host down for maintenance, et cetera. We took care of all of that on the backend. Also, we did all the patching and maintenance. So vCenter, the host, the storage, the capacity planning, making sure there’s enough hosts for the resources that are in the clusters. We took care of all of that, which is a lot of the effort that a lot of organizations go through to support their IT infrastructure. And that’s really where the value of cloud overall comes in. But when they go to something like an AWS or an Azure, they were having to flip the way that they monitored and patched and backed up and all of those pieces that we didn’t have to completely tip the apple cart because everything we were doing was based on VMware. So, we gave them a true cloud model with the VMware technology stack.

Scott Kinka:

So basically, chances are that the first virtualization that the customer did, and if they’re moving an on-prem workload, you had a very, very high chance that it was VMware-based already. You had an infrastructure with the license flexibility that you were discussing that they could, but they’re still shipping VMware,

AJ Kuftic:

VMware on their side, VMware on our side. They would replicate over, and nothing had to change.

Scott Kinka):

It was also great for those who kind of don’t have the full background. There was also great orchestration, even natively in VMware, around failing from preem-based infrastructure to cloud-based infrastructure. And again, that flexibility around licensing. So, all that aside for a moment, I don’t want to call it a utopia, but it became the world that we got accustomed to living in. Right? Yeah, absolutely. Let me ask you the question in two ways. So, all of these bundles, now that the 160 skews are left, I’ll ask the question two ways. One, how was that affecting the service provider market, who’s filled this very unique need around VMware and enabled VMware-based infrastructure as a service to be competitive to the hyperscale clouds that you just mentioned? So, was it affecting them? And then you could answer them in either order, and I’ll remind you so we stay on track, but I’ll obviously also ask you the same question from the customer side. How’s their world now going to change on a go-forward basis? And then we’ll get to how you guys see resolution, but let’s just set the problem up cleanly: service provider problem, customer problem, given recent changes.

AJ Kuftic:

So, from the service provider standpoint, it’s a shift for us. When you go from per gig of RAM to per core, those are two different models, right? It’s like selling apples by the bag versus selling ’em individually. It’s a how do you price those comparatively? So we have spent a lot of time, a lot of effort on how do we change our costs and how do we change our cost model and how we go to market to reflect that. VMware is also changing the way that we handle our licensing, where we are also being charged for hosts that are powered on in our clusters, even if they’re not fully utilized. So, for us, there’s also optimizations that we’re doing to make sure that our clusters are as big as they need to be and not too big because then we’re paying for licenses that are not being used. So, from a service provider standpoint, this is a huge exercise for us that’s going to take us probably six to 12 months probably to fully understand.

 

Broadcom’s VMware Vision: Reinventing Cloud Provider Relations

Scott Kinka:

What happens when somebody’s in terms of an offering that was built around a previous VMware assumption of the way costs? I guess as you’re doing that conversion, you’re having to sort of play the game on both sides and sort of work it out in a way that’s fair for the customer but also covers your costs. How are you guys sort of bridging that gap?

AJ Kuftic:

So that is where we have really smart people. What’s funny is I’m not one of ’em, no. We have incredibly smart people who have spent a massive amount of time. Here’s a spreadsheet of all of the customers. Here’s how much they’re all using. Here are all of our clusters, all of our hosts in the pipeline, and how do we actually manipulate the resources around to make it so that we are not burdening our clients because it’s not something they should have to think about? Got it. And that we’re not burdening our finances to handle this. And how do we spin this into a better light? So, for us, there are two things that we are doing. Number one of the actual good changes of this core from Giga brand to per core change is something that the service provider space has been asking for years, which is to fully bring your own licensing capability. So, if you have Cloud Foundation licensing as an on-prem subscription today, you can come to a service provider and bring those licenses with you, and we’ll provide the hardware, and then we’ll patch it all for you. So effectively, if you own these licenses already, you can just bring ’em to us, and we’ll take care of it, which is something we’ve been asking for a really long time, and it wasn’t known.

Scott Kinka:

You have customers who are like, I’ve made this big investment. I want to move to the cloud. I’m moving VMware to VMware, and I can’t use my stuff. It was not included in the enterprise agreements that service providers had to be able to do that.

AJ Kuftic:

And the big thing there was never that they couldn’t do it. It wasn’t like, oh, there’s literally no way possible for us to do this. It was that we, as the service providers, were charged per gig of RAM, and you were buying it per core. What does that correlation mean? Is it four gigs of RAM is equal to one core? Is it six? Is it eight? What does that actually align to? So now that we’re being charged per core and you’re being charged per core, and it’s the cloud foundation license for you, and it’s the cloud foundation license for me, you can bring that, and I think that’s a really, really big positive shift that comes out of this.

Scott Kinka:

Now, you mentioned two things. That’s one. What was the second one?

AJ Kuftic:

The second thing is that this also means that we are going to make distinct changes to the way that we deliver our private cloud model. So, we’ve had two cloud models for a while. We had a public cloud, which is super granular. You get a resource pool number of gigs of RAM, and you can grow super granularly, but some people need private cloud for different reasons. They want a VDI platform, so they need to have vCenter access. We did not provide direct vCenter access. We provided it through VMware Cloud Director in our shared cloud. So, we have clients who want per-host resources. They have compliance requirements where they don’t have the ability to share hosts with other clients. They want to avoid any questions about it, and they’re fine to do just their own private hosts, or you have database requirements. Oracle licensing and SQL licensing are still there. They’re just as coming for you as anybody else. So there were other reasons to do that. So what we’re doing is changing the way our private cloud works so that we’re able to match that per core model so that we can enable the full, bring-your-own licensing because it still has to go to dedicated hardware. You can’t bring it into our shared environment. So that’s really where we’re making these changes that I think are going to be beneficial. I think Broadcom has also made it pretty clear that they want to invest in the service providers, which is not something you necessarily asked about, but one of the challenges that we’ve run into is that VMware, in the past, has leaned heavily on the Azure AWS Google Cloud VMware story, even though there were 4,000 VMware cloud providers across the globe in many, many more regions than any of the hyper scalers were. So we had to go compete against VMware doing the exact same thing, but they were taking them to VMC on AWS because their sellers got comped a certain way, or it fit the story overall. When you are a publicly traded company, you want to have this true cloud story, and they really wanted to make sure that they weren’t getting let, the market wasn’t leaving them behind because they weren’t seen as cloud, but if they attached to all the big hyperscalers, they could get the kind of drag to go along with it. There’s definitely a reinvestment in the cloud providers and making sure that we are on equal footing. So they are changing not all.

Scott Kinka:

Or at least that’s what we’re hearing, right? Is that the case?

AJ Kuftic:

I mean, they have definitely been tiring out. The cloud providers. Expedient is at their top tier or will be at their top tier, and this allows us to have certain capabilities and certain go-to-market messaging with VMware. Their sellers will have compensation for selling things on expedient, which is something that has never happened before. And that is a big, big, big deal for us because it now allows us to have another vendor partner in the field. We’ve been successful with other vendors who will co-sell with us and who will use us as a replication target. So Zerto and Cohesity have done sales with us in the field because we provide value to their overall solution, right? Zerto needs somewhere to go, and we’re somewhere to go. Cohesity: you want to replicate your backups somewhere else, but you don’t want to buy a second cluster, or you don’t want to have a second data center. Hey, here’s expedient. They do Cohesity more than anybody else in North America. So this is a way for us to go to market with our vendors, and VMware has always been one that we’ve been trying to get them to see the value. You see the vision, and they’re finally doing it. So

Scott Kinka:

It’s a good news story at a high level, a good news story in your world.

AJ Kuftic:

It’s a mostly good news story. I will say that VMware’s costs were also passed to us. So, on our side, it’s making sure that we’re keeping our costs in line and making sure that our clients aren’t also seeing the same increases that they are. So, to your second question, what’s happening to the client side? Big cost increases. I have yet to hear a client say anything less than a 50% increase in costs. It is significant, and a lot of it comes back to what I find to be super interesting. VMware, before Broadcom, really hated the concept of shelfware. This is why VMware never really bundled things; they wanted you to use all of the pieces. Broadcom, on the other hand, very much would like you to bundle everything, and it’s up to you to get the full value out of what you purchased.

Scott Kinka:

Got it. So the increase, the 50% increase, is not based on, I mean, they were already paying per core on the base VMware license.

AJ Kuftic:

They were paying per socket. It was per.

Clients Contemplate Platform Shifts Amidst Licensing Overhaul

Scott Kinka:

Okay, got you. So that’s part of it, but the real increase is based on the fact that you just can’t buy it at this granular level anymore. So you’re buying the bundle, and you may have parts of that that you’re not utilizing.

AJ Kuftic:

And so you’re going from, I only cared about vCenter and ESX because that’s all I wanted. I don’t need vsan, I don’t need NSX because, for whatever reason, I don’t want to use those products. Now, they’re in the bundle whether you’d like it or not. And that shift beyond, so not only are you getting that bundling that comes into that of the vsan, the NSX, the Aria suite, all of that is now bundled in. So you now have this bundle that has actually, in my opinion, a ton of value inside of it, but you’re being forced to take it. You really get a choice. On the other side of this is that you’re going from perpetual licensing, where I paid the list price on Enterprise Plus because I know this off the top of my head, and it’s creepy that I was about $3,500 a socket. And so, for most two-socket boxes, they paid seven grand forest socket license, and then they paid about a thousand to 1500 bucks a year for support. You’re now going to that almost monthly. Got it. So that is another significant cost coming,

Scott Kinka:

So they’re going to pay you on an annuity, right? One, frankly, aside from being a box hugger or still depreciating gear, one of the big advantages of running it yourself is that you had that perpetual license flexibility, and you weren’t paying on an annuity. I mean, the annuity argument for keeping it local doesn’t even exist anymore, for all intents and purposes, based on what you’re

AJ Kuftic:

Saying. Yeah, you’ve moved everything. Just so we’re clear, though, everything in the world is effectively moving to as many monthly recurring revenue streams as possible. TV streaming music, everything that you do is moving to a subscription model because everybody saw like, oh wait, you mean I can just get paid monthly for this thing? Great. To the point where there’s now a subscription management software that will look at your bank account and go, oh, you’re paying for Netflix and Hulu and Disney Plus and all these other things and figure out what you could get rid of and help you unsubscribe. This is the same thing. The entire industry has been moving to subscriptions for a very, very long time. Also, if you look at the cloud, overall, the cloud has been a subscription forever. Literally, that’s the whole point. You’re moving everything from CapEx to OpEx. The perpetual to-subscription model is the same thing. Effectively, you’re moving from a CapEx purchase of licensing to an OPEX subscription, and so that’s where a lot of these other costs are coming from because you didn’t have that in your budget for the last five years because you were just paying for support. And if you were a large enough company, you had an ELA, and you signed it once every three years, you had your support all baked into that ELA, and then you were done versus now it’s a monthly subscription. It is something that is coming out every single month and the overall total cost is higher. I do think, though, that if you want to stay on VMware, you’re not entirely crazy, right? There are a number of different reasons to do that. You have an application that is certified to run on VMware and can’t run on other things without losing support from that application vendor. You have a ton of ecosystem around it as well.

Scott Kinka:

You’ve got staff trained.

AJ Kuftic:

You’ve got staff trained, monitor management, management capability.

Scott Kinka:

Totally. But lemme ask you this question quickly before we jump into the fun at the end here. We were running up against time. Sure. Are you getting asked that question as you’re talking to customers? Are they going? Well, I guess this is the time I should consider moving my workloads from VMware. Forget about prem or cloud or with Expedia and just in general. Moving off of VMware. Are they saying it is time for me to rebuild my workload on some other platform?

AJ Kuftic:

Yes, a hundred percent.

Scott Kinka:

Okay. How do you normally respond to them when they ask you that question?

AJ Kuftic:

I respond to them with, yeah, I totally get it. Have you considered backups? Have you considered Dr? Have you considered monitoring? Have you considered all the operational things? Have you considered what you’re going to go to? This is the personal opinion of AJ. This is not the opinion of Expedient as a whole company. Just want to make sure that we’re clear here. There are two options.

Scott Kinka:

Yeah, I get you.

AJ Kuftic:

There are two options that you can go to. One is Nutanix. They are very clearly the number two, and they’re the number two hypervisor. They’re the number one alternative to VMware. Number two is HyperV. If you were trying to stay on a traditional infrastructure stack, those are the two. You will hear Prox Max, Verge, scale, and KDM. All of those are fully capable of running the virtual machines. That’s not the hard part. The hard part is all the stuff around it. How do I back it up?

Scott Kinka:

How do you size it in every regard? Yeah, the applications, the workload itself, the maintaining it, monitoring it, having staffing for it, and frankly, any one of those downstream platforms are going to be harder to staff for. I mean, it’s one of the things to think about. There’s a lot of VMware talent out there. So I think one of the things, if I was talking to a client about an ROI and making a conversion on this one, I’d say, are you considering reeducation costs? Are you considering staff losses? Are you including staff rehiring? Are you including rehiring potentially at a higher rate? Maybe what you’re shifting to is in more demand because there is a little bit of this question mark around VMware right now and what people should do. So I mean, I think the headline is you’ve got to consider all of the, be honest with yourself about all of the cost components,

AJ Kuftic:

And I think there’s also a number of people who are saying, well, if I’m going to pay this much, I might as well just take it all to Amazon, or I might as well take it all to Azure. So that’s also the same thing, but the same questions are there. How are you backing it up?

Scott Kinka:

It doesn’t reduce by any stretch, going that route, right?

AJ Kuftic:

So I think there’s a lot of people who are looking at that overall shift. VMware has put a very large financial lever in place if you would like to pull it so you can leverage service providers like us. We went through and did our overall cost analysis of our clients and saw that we are only going to see a very, very small single-digit percentage change for our clients licensing-wise in regards to their overall account for most of our clients. It is something that, one, we’ve done because we’ve been able to optimize our things over the last 15 years, and two, we’ve done a really good job of keeping our costs in line. So we’ve been able to deliver this in a very efficient manner. I think the other thing to look at is from an alternative standpoint: if you’re going to look down that path if you’re going to look down trying to move off of VMware, think about where you’re going to be in three years. Think about where you’re going to be in five years. Is this something that is worth the shift and the move if, three years down the line, your investments aren’t where they need to be? So, I look at things like three-tier storage. If you’re a big pure shop, if you’re a big Dell storage shop, if you’re a big, I’m thinking of other ones. Hitachi, if you’re in any of those sorts of spaces, Nutanix doesn’t have support for that.  So this again comes into the overall ecosystem, the overall infrastructure that you have, and it might be more of a shift than just the hypervisor that your platform is running.

Scott Kinka:

A hundred percent. Yeah, great advice there.

AJ Kuftic:

So I highly encourage everybody to investigate it. I encourage everybody to talk to providers like us. We have done an HV as a service. We have the ability to back up an HV. We have the ability to do A-H-V-D-R, so we can leverage those things and assist as a service provider in helping you get your workloads to where you want them to be. This is something we’ve done not just for a VMware hypervisor shift but also when clients are looking at the cloud, making sure workloads land in the right place. That’s our whole cloud. Different sort of construct of you putting the right workloads on the right platform to make sure that you’re optimizing your spend and your time.

 

Unpacking the Future of AI

Scott Kinka:

Yeah, and certainly a theme that our listeners have heard quite a bit on this show, so they’re certainly familiar with that kind of approach, a place for everything and everything in his place. But I appreciate the frank conversation, agent. I appreciate you getting on so quickly with us and being willing to talk about this. Not every service provider has been willing to have the VMware conversation, and I think getting out in the open for our listeners and our customers is a really important first step in getting that going. So, I appreciate your transparency on that. I’d like to end with a couple, and we have left with a couple of fun questions if you don’t mind. So I’ll ask you two super quick ones and then maybe one that’s a little bit more heady. I’ll do ’em in this order. What are you reading right now? Is there anything that you want to share with our listeners or listen to? Not everybody’s an avid reader, but

AJ Kuftic:

Yeah, I’m not a big book reader. I read a lot of news. One thing I’ve been reading about a lot lately is the shift to EVs, and how that’s working how that’s being shifted around, and one of the more interesting things about Chinese EV makers, I find this in the entire Chinese EV market. Fascinating for some reason, and a lot of it comes back to pricing versus capability, where that investment goes, and then how that expands to the rest of the world. There’s a very large company called BYD that came out yesterday and basically said, we’re not coming to us. I found that somewhat interesting but also fairly logical in terms of what we’ve prioritized in the US investments into domestic makers, which is a good thing. But it means that when these other large companies that could compete in the market come in, they don’t really want to come in like we’re going to get pushed out or left behind because they want to prioritize these other things. So, I find that fascinating.

Scott Kinka:

There’s definitely the balance of propping up manufacturing in our economy versus ensuring fair pricing in the market, and you have all those pressures in both directions. It’s super interesting. Alright, the next pandemic is much worse, and it’s a zombie apocalypse, and all of our infrastructure breaks down. There’s only one application that remains working on your phone, whether it be your Android or your iPhone, or what have you got to choose before it happens? What’s the one application you can’t live without in that situation?

AJ Kuftic:

Ooh, I think it’s got to be either a podcast app or a music app.

Scott Kinka:

Got it.

AJ Kuftic:

That then means that there’s still a service on the backend that has the music, right?

Scott Kinka:

Exactly, you’re making the rules here.

AJ Kuftic:

Right? There are podcasts that are still coming in, or there’s music that’s still coming in, but I think that I’ve more or less cut most of my social media usage over the last two years down fairly dramatically. Doom Scrolling was a thing, and don’t do that.

Scott Kinka:

I love it. Don’t do doom scrolling if you don’t doming. Take nothing else from this episode.

AJ Kuftic:

I got it. Don’t doom scroll. It’s bad. It doesn’t feel good.

Scott Kinka:

This last one here, and you could get nerdy with this. We might end up back in tech, but I’m going to ask you for a shameless prediction: something in the 12 to 18-month window. It could be tech, it could be business, it could be sports, it doesn’t matter. Tell us what’s going to happen so we have something to look back on.

AJ Kuftic:

I think a lot of the early people running into AI are going to fail.

Scott Kinka:

Okay, tell me why.

AJ Kuftic:

I think there are a lot of people who, I mean, just thinking back in my career, I go back to 2009, 2010 when the cloud was a thing, the cloud was the biggest thing. If you’re not in the cloud, your company is dead. Guess what? It’s 2024. We’re all still here, and 70% of the work still runs. So maybe that didn’t pan out so hot. And that’s because I think a lot of people missed what the value proposition was. They just said, oh, if I go to the cloud, I’m going to go faster. And that’s not really the case. You’re spending your money monthly. But I think the way I see AI right now is there are a lot of people who are looking at AI. In fact, it was the CEO or CFO of OpenText who came out yesterday and said the solution to the talent problem is less talent, and that is a despicable take, in my opinion. You cannot get rid of the humans in AI.  AI does not replace humans. What AI needs to do is enhance humans. It needs to be the extension, right? So I like to pick on one of my coworkers who’s an SME on our cloud platform who probably spends two hours a day answering instant messages of questions that are in our documentation. If I can put something in place that allows our users to ask questions and get real honest, actually correct answers from our knowledge base in a conversational form, I just saved my SME two hours of his day. I saved a quarter of his year from answering these questions. So that’s a big deal. A lot of the things that we’ve built internally, we’ve built things like a ticket summarizer that looks at our ticketing system and takes all the notes and all the changes and all the things and summarizes the ticket into a human-readable form. It’s really great. I have found that AI can be really, really helpful. My problem with AI is usually that it’s wrong. My superpower is that I’m not bad at writing. I can write things, and I can make things sound good. I can talk, make talk tracks, and all that sort of stuff. As a somewhat wordy and creative person, I can do those things. That’s not the hard part. It’s getting the answers and getting them right. And so for me, on a lot of occasions, I end up rewriting or having to just go look up the right answer for what was actually in there. So I think a lot of people are running out there. There’s a story that just came out a week ago about Air Canada and how the chatbot on their website made up a bereavement refund policy. And when the person called in to get their refund, the company was like, no, we don’t do that. And they’re like, yeah, but your chatbot on your website said you do. And Air Canada was like, yeah, but that’s the chatbot. That doesn’t matter. And a court said, no, you have to refund them their money. That’s a refund policy. This is the sort of thing that I think companies don’t understand is going to happen, and it makes you look silly. I don’t think that AI is a bad thing. I think it’s a truly great thing and a truly great technology, but it needs to be applied correctly. Just like with the cloud, just like with any other technology, when we apply it correctly, it will work out better for you. If you are just ChatGPTing, everything. The cadence looks wrong, but the words look good. It’s just a really confident, incorrect person on the internet, and we don’t need more of that.

Scott Kinka:

I love that ChatGPT is a really confident but incorrect person. I love that. That’s a hot take right there. And obviously, we’ll put it in context. This has been a lot of fun. Forty-five minutes disappeared very quickly as we were going through this, and I appreciate your transparency around the VMware stuff. As I said, we don’t have a lot of people who want to jump right on this grenade, but we’re getting a lot of questions about it. Absolutely. So, I thought it was super useful. I hope you had a good time. I know we did.

AJ Kuftic:

Absolutely. I’d be happy to come back, and if anybody out there has any more questions about how they can mitigate their VMware spending, are looking for their VMware platform, or want to talk about alternatives, let us know. We’re here at Expedient. We’re happy to help.

Scott Kinka:

Fantastic. And for those of you who work with Bridgepoint Today, that’s Expedient. It is one of our top partners in the space, and we’re committed to working with each other on these types of things. We’re very involved together, and hopefully, we can help broker some of those conversations. Looking forward to having them. Thank you all for spending the time to listen to this episode, and we will catch up with you soon on another episode of The Bridge.