Hybrid Cloud is Fools Gold - CTO044

Machine-generated from the episode audio and not hand-corrected, so names and technical terms may be imperfect. The audio is authoritative.

All right, welcome to episode 44 of this TTO Advisor podcast. It's been a little over a month. I thought I'd do these with a little more frequency or at least stability, but you know what? It's a free podcast so you're welcome anyway. So on the line I have the second time Tyler has been on. Tyler Britton. Tyler, can you go ahead and introduce yourself? Yeah, sure, no problem, Keith. My name is Tyler Britton. I work for IBM and IBM Cloud in the office of CTO, focusing mainly on BlueBox, IBM's managed OpenStack offering.

So Tyler, you just got back from a bunch of traveling. What's the schedule like? You were just telling me right before the recording. What's the travel schedule for the past couple of weeks and the next few weeks? Yeah, it's crazy. I mean, you know, when you travel and you do conferences and things like that, you kind of, in your mind, in the year, have it like picture to spread out. But just the way conference scheduling usually works is there's a lot in the spring and a lot in the fall.

So there was a little bit of a low, but then this past, say, month and a half or so was in Germany for the Cloud Foundry Summit. Then I went to a Munktoberfest up in Maine, which is one of my favorite conferences. And then last week, I was at Cloud Expo Asia in Singapore, as well as doing some customer meetings in Jakarta and Indonesia. Now I'm home this week, and then next week is the OpenStack Summit in Barcelona. Wow, man. That is a rough, rough schedule.

So I did the math for last week. From Monday through Friday when I left, I was in an airplane for 46 hours of the week. Oh, Jesus. I cannot do that. You know what? I'm a tall guy, so unless I'm flying business class, and even flying business class, that is still a tough, tough schedule to be in an airplane. It was pretty brutal. I will say I flew out Qatar Airways for the long legs, and that was a pretty nice airline with pretty brand-new planes, and the seats are a little bigger.

So even in that economy, I wouldn't volunteer for it, but it wasn't as bad as it could have been. So when you say you support the OpenStack offering over at BlueBox, what exactly is your role? Is it more of an evangelist type of role, a developer advocate? What do you do for BlueBox? Yeah, I mean, that's pretty much it. We call it advocate, technical advocate. Other areas call it developer, evangelist, kind of that area of a mix of whether it's speaking at conferences, building content, demos, and blog posts, and stuff like that, as well as just straight-up customer meetings.

And it's starting to grow more also as far as BlueBox is integrated into the IBM Cloud Platform. Recently, it's been a lot more not just BlueBox, but BlueBox and Bluemix kind of all together. At some point, we got to get John, Amy, and Matt to have you on Geek Whisperers, so you can kind of tell the behind-the-scenes story of what it means to be a developer advocate and kind of those ups and downs. Because, you know, from the outside looking in, that's an awesome job.

You get to spend 46 hours on an airplane, and you get not to be at home, and it's just, you know, you get all this foreign money in your pocket. It's got to be a great experience. But, yeah, the job doesn't sound for the weak of heart, and I'd love to, on another medium, just hear kind of what it's like behind the scenes. Oh, yeah. Yeah, I mean, while those are the bright parts, but then, yeah, it's besides the travel grind and stuff like that at times, there's a piece of it that, so I have my background, you know, when I was at EMC, you know, being a V Specialist and stuff like that, and even before that, working for a bar or in IT is, you know, you see things more kind of beginning, start to finish, whereas, you know, that type of role, it's a much shorter interaction with customers and people, where it's really, I mean, I'm sure you know, you know, being in it every day is, it's really satisfying to see stuff go from idea to finish, whether it's a, you know, new software rollout or something, you know, that's really impacting the business, to see it go from beginning to end.

So, I would say that's a piece of it that's rough, is you miss out on that component. So, it's, we're gonna have an interesting conversation today, because I think, you know, me and you on Twitter, we can, sometimes we fully agree, other times we kind of go back and forth on where we think the industry is going and what's important, and then other times, like, we're just way off. Today, I think we have a topic that we agree upon 100%, which is that, and I think this might even be the title of the podcast, that hybrid cloud is fool's gold.

Yep, 100%. And that may shock a little people. I mean, you work for Bluebox, and you do managed OpenStack, so there's kind of, traditionally, there's been kind of a managed OpenStack hybrid cloud story that we'll get into. So, that may surprise people. And then, in the past, I've always been this proponent of hybrid, or hybrid, more specifically, hybrid infrastructure, and those things are not exactly the same. Hybrid infrastructure and hybrid cloud. I wanted to get into that conversation, but before that, I think it'd be great for us to kind of walk, walk through and see if there's any areas that me and you don't agree on in this, in this whole journey of how we think hybrid cloud is fool's gold.

The first piece is, okay, let's talk about public cloud. Is public cloud a thing in the enterprise? Yeah, I mean, I think it is. I think it's in varying degrees. I don't think it's as widely deployed, kind of, as sometimes people, you know, I think there's, there's, there's decent pockets of it, but it hasn't fully taken off. When you look at, you know, Amazon's, you know, Amazon being the big player there, and you look at their, their AWS revenues, and they're really impressive for that space, but in comparison to the total overall, whatever we'd call it, infrastructure spend in the enterprise, it's, it's a rounding error still.

So that means there's a lot of stuff that's not in the public cloud. And the, and there's solutions for that, and I think Blue Box steps in and, and helps with some of that stuff that you don't want in your data center, but you need in a cloud-like environment or a hosted environment. And then there's, and I, so I agree with you. Public cloud is definitely a thing. Adoption in the enterprise is questionable. The, the two announcements, the one with the, we'll just call it VMware Cloud Foundation, I don't think either one of us called that cloud exactly.

That's, that's a, that's a thing, and there's use cases for it, but that's not, you know, cloud proper. And then there's the term private cloud. Is private cloud a thing? Yeah, I think it is. I think people, you know, the, the idea of really behind, to me, what cloud, and, you know, if you want to bring up the NIST definition for the millionth time, but to me, the main thing is, hey, it's, it's on demand. It's users are self-provisioning through an API or portal.

They're not opening tickets or talking to anyone. And from the user's perspective, it's relatively elastic. So they can ask for a bunch, you know, they can ask for a million VMs, but guess what? You can't do that on Amazon either. And, and the idea that like, well, only public cloud is really cloud, that you hear sometimes, I disagree, because you can go to Amazon and ask for, you know, depending which, you know, types of VMs or instances you ask for, you can get it coming, you know, you can get a message back saying that, no, I'm sorry, there's not enough resources available to do that in that particular availability zone or something.

So it happens. So the idea that it's this, cloud can only be this infinite scale thing, I think is a, is a, is a misnomer. So I think you can have private cloud, but I think what, what gets people kind of antsy about that term is people call a lot of things private cloud that are, you know, forget even just being, you know, hey, we're running some, some stuff on VMware. So we have a private cloud. And I think that's what gets under people's skin.

Yeah, and I agree. I think there's kind of two extremes. It's the people who say that, obviously VMware cloud, VMware, not cloud, VMware vSphere infrastructure by itself is not private cloud. But at the same token, you don't need this whole NIST definition of what it means to have private cloud. I think a company like Dropbox, who just moved from AWS to their own private data centers, I'm certain that a lot of the methodologies that they used at AWS, they still use a lot of those methodologies inside of their own data center, and they call that private cloud.

I think, yeah, absolutely. I think that's the key thing is kind of like the how it's consumed and operated, you know, like, that's, I'm sure the Dropbox people are, you know, getting API's from their internal team that they're, you know, plugging the software into. And to me, that's, that's the real key is we have customers that move to blue box off of, say, AWS, and they're using it in a similar fashion. And I think that's really the key, the key piece of it.

So on on premises solutions, I think one of the things that I've run into, and I think this is going to start to get into the hybrid cloud conversation, is that enterprises are not using private cloud in a way that vendors and pundits, and I'm, I'm one of those pundits years ago, who said that, you know what, you're going to replace your, at some point, your ERP is going to run on private clouds. That makes absolutely no sense. The legacy applications are going to run in legacy infrastructure, if you're going to have it in your private data center.

And you're going to new stuff that is kind of digital transformation that the architecture makes sense for private cloud, you're going to run that stuff on whether it's managed in house managed OpenStack, OpenStack, you run your own. The curse word that we both like to use in vCloud director, if you're going to run it in that, but something that you're going to build some type of pod that you're going to build, and you're going to run it on that, but the most of your data center for traditional applications are going to, is going to run on traditional infrastructure managed in a traditional fashion.

Do you agree with that statement, more or less? Yeah, I think I think what's really interesting is, at Munktoberfest, Brian Cantrell of Sun fame, and Joyent, and now Joyent slash Samsung, was giving a talk about kind of the oral history of and maintaining oral history of computers and stuff like that. And it was a really interesting discussion, talking about, you know, kind of how we got here. And then it turned into this really good discussion at lunch talking about, basically, the net net of it was how much old crap that was written, you know, 3040 years ago, that's still running today.

And kind of, you know, what what his hypothesis on it, which I agree with is, we have this idea that software is very ephemeral, like, we're gonna write this, we'll do it quick and dirty, it doesn't matter, because this thing will be replaced by when someone rewrites it in five years on the next new magic widget thing. So hey, we did a bunch of stuff with VMs, when containers come, people totally rewrite it. And it just, there's not a business case a lot of times to just rewrite things for the sake of rewriting them.

So stuff sticks around. If it works, it works. I mean, how much stuff is running on mainframe still, and those are, you know, stuff that was probably written 40 years ago, that's still running stuff that was written on alphas and, you know, and on and on and on. There's no real value to pull it forward and rewrite it into quote, you know, cloud native, if there's no business value behind it. So why do it so there's gonna be some, some percentage of the environment that's the same old stuff.

So just dealing with that is key. So and I think that is one of the disconnects that we've had in the industry at large is whether we've hyped up private cloud to be this methodology to run your entire data center, your entire infrastructure, and everything is going to be in private cloud. And hence, you're going to have this use case in which there's stuff that you want to run in your private cloud, that and then stuff that you're going to want to run in your public cloud.

And there's a integration, a single point of pain, a single pane of glass, which I think is the part that we're calling the youth to borrow the term, but well, the bollocks, the why you're not going to have a single pane of glass to manage your private data center in your public data center at one time that the or your public cloud, I want to say private, public data center, because that I think that's a thing. But the this this whole method is consuming infrastructure, this cloud methodology, there is no hybrid version of that.

Not at scale. Yeah, well, I mean, that's like that old, you know, what was the early days of public cloud and, and private cloud and stuff was the magic cloud bursting, right? Hey, we're a retailer, we're gonna we have all these websites, but it's Christmas. So we're gonna burst out to the public cloud. Yeah, who does that? Nobody does that, like, get out of here with that. One of my favorite examples was one of the I don't know if it was New York Roadrunners, or one of the organizations that that runs one of the marathons, they actually did the opposite.

They would take all of their, you know, they had an in house data center where they ran stuff year round. And then when it was coming up to marathon week, they would move or shut off or move all their normal systems elsewhere. And the only thing running in their data center was the timing and website reporting systems and everything for the marathon. Everything else was all the all the high value stuff was actually running inside the data center. And everything else either was shut off or running somewhere else or on Yeah, on Lord, that's, that's actually because I mean, think about that.

If you're if you're a retailer, if you're saying your target or someone, you know, the the winter, you know, this is the holiday times, your critical time of the year, you're like, yeah, I'm gonna we're gonna burst this this stuff out to infrastructure we've never we don't normally use during the year, we're gonna introduce a whole bunch of new stuff this time of the year, when we burst out to some provider, and that's just going to magically happen. So it's always a thing that was kind of talked about by vendors, but no one really did.

I mean, the idea of burstable infrastructure companies that are doing that are basically running that entire workload on say, a public cloud, where they have the spare capacity burst it out. And, and that's one of those perfect use cases for public cloud that makes sense. But I think to your point that that notion of we have these two different clouds, there's this magic interface where I can drag and drop stuff and move them around. And not only does it like, fall apart, as soon as you start kind of the Socratic method kind of questioning how you're going to actually do that.

It just it just doesn't match up with the reality of what people have in their data centers. Yeah, and I think this is one of those things that people are might hear this conversation between us for the first time and kind of shaking their head. Like, I don't agree. Usually we agree with Keith and Tyler, but they're off on this. So let's walk through this just a little bit. So let's talk about the types of workloads that you will have in a private cloud, and what that operations looks like.

So, you know, you there's, in my experience, the driver for private cloud is some business group. So let's walk it back from the business. There's a business group that wants a application, that for whatever reason, that application needs to be ran on premises. And they want a API, they want a cloud API to it. So that kind of initially drives the business case for building a private cloud. The business case normally doesn't come from IT infrastructure saying, Oh, you know what, this way that we've been running infrastructure for the past 30 years, we want to do it differently in a model that's ephemeral.

And that doesn't naturally fit that, that that's usually not the case. So when we get those types of requests, those are normally either you're going to take like a Comcast perspective, and you're going to build a big general purpose private cloud. Or you're just going to say, you know what, I'm going to go to HP or IBM or one of my favorite vendors or integrators, and I'm going to have them build a pod for this application. And I'm going to set this pod in my data center, and this group can consume it however they choose to consume it.

Yeah, I mean, I think I think you've hit on the key thing is to me true private cloud, if you will, is consumed in the same way as public cloud, it just, you know, kind of where it's, it's the difference of single tenant versus multi tenant, versus physical, you know, in physical location. So for you, like I said, we do we do manage private clouds in the customer data center or in one of the IBM software data centers. So what are the drivers that bring customers to that is, sometimes it's data locality, right?

distance the data. So they have, you know, talking about those traditional systems, they have some main line of business application that where all their data is could be petabytes of data, and they're writing some new app to talk to it. So say they're a, you know, a hotel chain, and they have this old school Unix, you know, reservation system, but they want to write a new web mobile app, they want to write a cloud native, but they want to have it close to the data sets, they're going to be hitting.

So like, that's one use case, data sovereignty issues with governments and different countries. And those types of things are generally it there's generally not a technology reason, other than that data locality piece, most of it is some outside reason. But it's, hey, we want to consume this in this manner, whether it's public or private, but some outside reason is forcing us to run it in a private manner. So I think that's really the main driver. And that's what to me really differentiates between, hey, are we doing a real private cloud?

Or are we just having a bunch of VMs and we're calling on vSphere, and we're calling it private cloud. Right. So and then the next step, the next step in the hybrid cloud journey is to say, okay, I have this private cloud. And I have a use case where this private cloud needs to not only communicate and integrate with public cloud, but I need to manage it in the same fashion in which I manage it on premises. And I think that's where the argument starts to fall apart.

Where does that really ever happen? Um, so I've seen that the use case of, hey, we're a group and we have whatever, you know, private, public, or some app that we're developing cloud native, and there's some mix of public and private cloud. To me, that's not super new. And that's one of those things that happens, you know, if you want to call it a hybrid cloud or whatever, but it happens the layer above. So if I'm a, you know, shop that's running an app, and I'm however I'm deploying it, let's say I'm using Ansible, or Terraform, or, or whoever, you know, Urban Code Deploy, or any of these tooling, that's generally like, oh, here's another place we can point the tooling at.

Cool. Like, so customers we've had that have been on AWS that have started to use Bluebox, the discussion with the developers is pretty short. They're like, okay, cool. What's the API? Oh, oh, there's a plug. Oh, there's a module for Ansible. Okay, this is how we ask for VMs. That's sort of like this, okay, we get it. And then everything's fine. So there's really not that big of a deal. I think where it really gets into, you know, true bollocks territory is at that infrastructure layer, when you're talking to infrastructure, people that want to build a cloud, that's already pre-connected and integrated in single pane of glass, and then hand it to that next layer up.

It's just like, it's not the layer where you make that interoperability happen. Yeah, and I think and so first, I'm like, wow, I was getting excited. Like, I got an area that I don't agree with Tyler with. But unfortunately, no, we agree, which is that infrastructure driven integration that, you know, one of the ideas both of me and you have seen products that says, you know what, when I don't have capacity, the infrastructure decides when to send the workload to a public cloud provider, based on some type of golden set of catalogs, we the infrastructure team decides where the workload runs.

Maybe there's an argument for that. Maybe. And that's a that's a strenuous argument to make there in those places. Yeah, and I don't believe I don't believe in the I think the control should be in the app team. I think the infrastructure team probably has a and this is where we get away from the argument for hybrid cloud into the argument for hybrid infrastructure. Because I think what you described is hybrid infrastructure, where from a application perspective, from a infrastructure team perspective, there are going to be workloads that run in the public cloud, no doubt, there's going to be workloads that run in the public cloud, that needs to communicate back to those on premises solutions.

There doesn't need to be a infrastructure, a infrastructure layer that provides a cloud gateway to those to a cloud interface to those two pieces of things that that's all stuff that can be done at the application level layer. There's no reason for the infrastructure team to recreate that. That interface between those two. Those two environments. Yeah, absolutely. I think that's the the whether you want to call it in the application itself, which in a lot of times, you know, that's kind of a misunderstanding.

I've seen infrastructure people already say cloud native apps that apps handle this stuff is yes, sometimes the app is making the AW, you know, the EC to call the spin up more instances, the app itself. But a lot of times, it's just the tooling that the app team's using to deploy and manage. So if they're using some CI CD system where they push a new bit of code, and it's going to wipe out and replace the Docker containers that the that the you know, that version of the app is running into a place with the new version, you'll have that piece of it, that tooling is generally what's that can talk to multiple infrastructures.

And I think you have a key point is, it's about providing those capabilities to the developers. And that's where I think infrastructure teams are wasting so much time to try and make this magic, useless hybrid cloud thing with a single pane of glass. Whereas I think if they actually talk more than developers, developers would kill for, hey, we wrote an integration layer that allows you to make API calls into this old mainframe app, it converts it into, you know, the proper assist calls and stuff like that, like, oh, wow, then we could write, you know, when we write our code, we can just call this, you know, REST API, and then talk to this old system like that, I think is where the infrastructure teams can do more integration to bring value to the app team than trying to make this magic hybrid cloud.

And I think where a lot of the confusion comes in for infrastructure groups is how do they get their arms around the infrastructure component of public cloud and private data center integration, so that hybrid management of the overall infrastructure. There's pieces of data and there's metadata and services that the infrastructure team still needs to manage. They, while developers may call and consume public cloud APIs, those calls provision infrastructure that the infrastructure team has to account for from a configuration management, etc. piece.

And I think this is where, man, you get into a lot of debate, and I don't know if it's really debate. It's, I tell you what the infrastructure teams and enterprises are looking for, and you, and I think rightfully so, you say they're looking for the wrong thing. What they want is right, but how they're going about getting it is the wrong way. Hence, the, you know, the cloud, the VMware's cross cloud, where they take a layer two network and slam it into AWS.

That fixes some short-term problems for the networking group, but long-term it breaks how you consume AWS. So how do you consume Lambda and you, and you're trying to control the layer two of AWS. Those two things kind of don't, those two things will come to a head at some point. Yeah, well I think, I think it hits on, I was about to bring that up and I'm glad you did, is the networking piece. And to me is a lot of the kind of SDN magic that, that it's a similar thing to hybrid cloud.

It's more fool's gold that, that enterprise teams are buying into because it's something that they're comfortable with and used to. And the idea like, well this is our data center subnet, so we want to stretch this subnet so some of the data center stuff can run there. And you're like, wait, we fixed this thing years ago, it's called layer three, like routing is fine. There's no magic of being on the same subnet is, doesn't do anything magic for you, you know, besides, you know, forgetting broadcast domains and crap like that.

And they're specifically not designed to be encapsulated into layer three and stretch really long distances because it has no concept of bandwidth or latency or anything like that in layer two. So it's just like really and generally a bad idea and we really start to break it down. There's all these, well, hard-coded apps and really, really, really, and you start asking questions, you really break it down. Most places do not have that problem where they couldn't do it. It's just the, the bigger problem is, I think, where the real kind of, kind of disruption is happening right now in enterprises and IT in general.

It's not cloud, like we're just renting servers by the hour instead of paying, or by the month or by the minute instead of by the year or, you know, three years. That's really what cloud is. Yes, there's an API for it and all that kind of stuff, but that's really, it's not really a huge tech change there. What, what to me it really is, is it's this new methodology of how you're operating things when you want to call it agile and a lot of focus on configuration management and, you know, this type of new approach.

I think that's really what's breaking things down. And if you really get to, even from a networking perspective, hey, let's say we're met, you know, those, that, that Oracle, old Oracle app that's got a big Oracle database on the back end. It's a whole bunch of Java app servers and web servers in front of that. And, you know, if you think about those web servers and app servers are, are, are ephemeral. We could just pick them up and run them in a cloud, whether it's private or public.

Well, why not? Well, oh, we're gonna break it. The networking breaks. That's why we want to do Stretch Layer 2. Like, well, wait. Why? It's like, well, well, there's a JDBC connection on the back end. Well, all you have to do is change the config file on all those hosts and it's done. It's like, oh, well, people manually log it into servers. And like, that's where it gets into the, whoa, whoa, whoa. Why can't you just update it in Chef or Ansible or whatever and push these things over?

And I think that's really where the breakdown is. So, to your point, it's, it's kind of a, it's a band-aid fix of like, well, hey, we'll just stretch the Layer 2 and then all the problems go away. And it's like, well, no, you're just, you're prolonging your problems. You're trying to stay in your old mindset. You're trying to move into the cloud while staying in your old methodologies and mindsets for too much longer. And then we eased kind of that demarker for what type of applications are best suited for the public cloud versus what applications are best suited for the private data center.

If you have those concerns, that, the application that you're trying to stick in public cloud doesn't belong in the service that you're putting it in. There's other solutions for what you're trying to do and those solutions are called VMware Cloud on AWS and VMware Cloud Foundation and Bluebox. Those are, those are infrastructures and services designed for those types of applications. The stuff that you, you're putting in AWS proper or in an OpenStack driven API environment, those types of applications talk IP. They don't have these layer two broadcast issues.

You're not trying to run multicast. These apps, what you need to learn, what we need to learn how to do from an infrastructure perspective is integrate into the AWS control panel and have our enterprise tooling talk to those solutions so that we can get the reporting and the monitoring that we need to get out of that. We need that API driven data center from a monitoring perspective. We, you know, we may not be looking at cloud, at hybrid cloud, but this whole ideal of hybrid infrastructure is putting around the operations needed to understand, properly monitor, and deliver the value-add infrastructure teams still deliver because infrastructure teams still deliver value in the cloud, but using the native tools that's available to us from these various cloud providers.

Yeah, absolutely. I think that's one of the biggest misnomers about cloud and even serverless is this, this stupid concept of no ops where it's like, well, we don't need ops people anymore because the devs just push the thing and then the thing just runs and then you're kind of, ops is, Amazon's doing ops for you. It's like, well, no, from the second you could, you know, commit on that code till that, you know, till that app gets turned off for the last time, everything else is ops.

And it's like, there's still a ton of value to be had there, you know, monitoring and troubleshooting and managing the capacity expansions and all that kind of stuff still needs to be done. I think to your point, that's where the skill set needs to update for the infrastructure teams to really get comfortable in these, in these new worlds to be able to handle, you know, where, I mean, think about back in the day when everything was on-prem and then the companies got their first SaaS app, right?

They got Salesforce or something and IT struggled it. Well, can we integrate our, can we integrate LDAP? Can we, oh, well, how do we manage user? Like, how do we, like, how do we get our arms around when someone calls and it's broken? Like we just don't. And then they worked it out. They figured out kind of what the model is. And I think that's what orgs need to do is kind of work this into their model. And to me, that's also where I think we get, we give people kind of, we get excited about cloud and we get people kind of going the wrong direction.

And I that's why I'm not a huge fan of the whole pets and cattle analogy. Just because it, the way, not, not so much the analogy itself is not bad, but the way it's often used kind of puts this mindset of like, well, you have to rewrite everything to be cloud and magic and all this stuff. And really what it comes down to is the stuff that fits on cloud is stuff that's kind of ephemeral and repeatable. And that didn't just show up when cloud showed up, right?

They said, we think about web farms, web server farms are ephemeral and repeat, you know, repeatable and Java app servers. And we have tons of stuff. And I mean, think about even some stuff that you have clustered where it's, it's kind of an older app, but you have two or three of them running as long as one of them is up, you're, you're fine. Like, why couldn't that run on Amazon? So I think that's also kind of a mindset change for these infrastructure teams of don't just think about, well, it has to be written in, you know, node JS and running Docker for it to be cloud, you know, that, that cranky old Java apps that you hate updating.

Well, here it might be an easier way. The app team, instead of, you know, kind of manually hand-holding these servers and pushing, you know, application versions out, why can't you load that into a Docker container or, or run it on Cloud Foundry, right? Oh, well, this, this Java app's just a war file. Like, well, why can't we just run that on Cloud Foundry? You know, and thinking about those things. And I think that's really, it's a, a change in my, and I think that's, that's really the, the thing getting infrastructure teams to accept that hybrid cloud and this one pane of glass is not a real thing.

And that means instead of waiting for that to be a thing, so that way they can kind of check off the box that that's done, is just start learning more about how this other stuff works and integrates. And, and to me, I put it more on even on leadership in these companies. Like, are your, are your infrastructure teams way overworked? Are they, are they getting training? Are they going to classes? Are they reading blog posts? Do they have a lab?

You know, they, they, they need to get comfortable with these, these technologies and they're not going to do, there's no one week class you can, cloud class you can send them to. So I think, you know, it's, it's gaining the capabilities and access because there are plenty of, it's not that there aren't super smart people, you know, you just have to give them the opportunity to, to get up on this stuff. And I think, I think that's the pieces where enterprises are really getting right now is realizing that they kind of have a, you know, a, a, you know, skills deficit and that's really the problem.

And they can't just hire them. So, you know, I can't put a monster ad for 10 years of doctor experience and sign people up. So I think, you know, they're starting to learn that they can level up their people and get them access to these things and really start to accelerate this, this transition to this new operational model, less, less than it is, you know, cloud transformation. So Tyler, on that note, I think we'll wrap it up. I really appreciate you jumping on last minute to, to talk through this.

I think it was an exciting topic, something that is probably going to drive the industry for the next few years. I'm going to Dell world this week, and this is going to be kind of one of my pokes at the Dell execs, which is how did you help companies transition to, you know, Dell EMC world and, and VMware? How did, how, what's the strategy for not shoehorning private data center stuff into public cloud, but really helping with some of these core issues that you, you helped bring up?

Where can people find you if they want to stalk you online? com. I actually had one talking about some of these issues specifically around the skills gap and kind of cloud disruption. So it's worth checking out. So I'll make sure to link that in the show notes. Of course, you can find this podcast in iTunes. Please rate us on iTunes. com. You can follow me on Twitter at CTO Advisor on Twitter. Talk to you guys, hopefully right after Dell world, Dell EMC world with some great commentary.

Thanks.