The Cloudcast #129 - Cloud Chaos, HyperConvergence & Commoditization

>> Cloudcast Media presents from the massive studios in Raleigh, North Carolina. This is the Cloudcast with Aaron Dell and Brian Gracie, bringing you the best of Cloud Computing from around the world. >> Good morning, good evening wherever you are and welcome back to the Cloudcast. Just myself tonight Aaron is off in Seattle. Before we jump in tonight's show, I want to give a huge thank you to our audience, who we consider really kind of the best tech community in the world. We were a little behind as a few weeks ago. You guys came through, you stepped up big, we've exceeded our $5,000 goal for the Krispy Cream run for the North Carolina Children's Hospital. So from Aaron and from me, huge thank you to everybody. We'll be posting lists of donors and all sorts of stuff, but thank you, thank you so much. So first I do thank yous and now I've got to do apologies, because tonight's guests are two guys who have been great friends of the show, friends in the community and two people that Aaron and I have been incredibly slack about having his guests on the show. So Greg Nearman from Speaking in Tech and Stu Miniman from Wikibon, guys welcome to the show and our deepest apologies for taking this long to finally get you on the show. >> Well, I guess it's a pleasure and an honor now, right? >> Yeah, absolutely. So I can't react to any more about it. >> We're not, we're not very good Southern gentleman that we haven't invited you on any earlier than this, but thank you both for coming on tonight. >> So I still gladly can finally do it, Brian. >> Absolutely, so, so you know, I was thinking about this, because one of the things we were going to talk about tonight was sort of like, you know, disruption and things changing and commoditization and stuff, and I was thinking, you know, I don't, I don't know a lot of little kids these days who have the name Brian Greg or Stu. So just ourselves, we're becoming sort of dinosaurs from a naming perspective. >> That's an interesting observation, I never thought about that like that. >> You got all sorts of, you know, everybody's a little kid has to have sort of a common, you know, kind of a unique name. We used to have lots of Dave's and Mike's and Greg's and John's and Stu's, but they just don't exist anymore. >> I would have changed my name to Tyler then. >> There you go. >> Or Forest or something. >> Forest. >> So let's talk about, let's start by talking about disruption. You guys both, you know, heavily involved in the enterprise space, in cloud, in, you know, in tech, but, you know, you also kind of are involved, you know, with what's going on in disruptions, Stu, you weren't necessarily out there, but you were covering the open compute project, which is sort of becoming like the gigantic open hardware project. What's going on? Is all hardware completely commoditized at this point? Well, I believe all hardware is now owned by Lenovo, so there's that too. You know, they picked up IBM 686, and now they're picking over the, picking up the Motorola stuff from Google, so yeah, absolutely, it is really interesting times for the hardware business, and, you know, especially being an infrastructure guy, it's fascinating to watch how some of these disruptions are coming to the market, and, you know, some changes happening fast. You said, open compute this week, it was the fifth of these conferences, Zuckerberg himself was on stage, talking about some of the things there, and, you know, who would have thought just a couple of years ago, you know, Facebook is an important player in infrastructure discussions, so, you know, it just kind of blows our mind. So Greg, you know, you keep it with us a lot, your company, my company, we're both sort of to a certain extent in the hardware business. Are you seeing this trickle down at all yet into the enterprise, or are these just, you know, big shows for the WebScale people at this point? Well, you know, I've got some opinions about that show, and to me, in my opinion, it really is a great way for Facebook to get everybody to standardize on designs so they can buy hardware cheaper, and the community at large benefits from that, I suppose, but I don't see many companies, enterprise companies really buying those at scale. There's a lot of commodity servers don't get me wrong, but to Facebook design specifications, I don't see it that much, I'm sure it's happening, but it's common. I would totally agree with you. If you look at, you know, who this benefits today, it's Facebook. Microsoft made some interesting news that they actually gave a design that they had applied for a patent to open compute, so Microsoft, obviously, for a viewer, wants to commoditize hardware too, but it's the WebScale, HyperScale guys, maybe some of the big financials, and it will take a long time for it to even trickle down a little bit to the enterprise when we're talking about the OCP servers themselves. Yeah, that's what I'm trying to figure out. I get everybody who wants to compete with Google in essence, right, because you're competing on a penny's per click and penny's per kilowatt, but what's the break-even number? How many servers do you have to have before this makes sense, or how many servers do you have to buy from someone like Quanta? Is it two shipping containers full? Is it 10? Can you buy a half a shipping container? Where does it start? That's the piece I'm curious, and I never really hear that discussed anywhere. Yeah, and Brian, it's actually something that I've been keeping an eye on. If you talk in the networking space, Whitebox Switches, what Cumulus and PK-8, some of those guys are doing, it's not only how many you're going to buy, but how often you're going to need to grow and update your environment. So service providers, great for them to do this. Enterprise that's going to buy 10 Switches today and not upgrade them for five years, probably not going to happen. So that rate of change in the growth, if you're not deploying racks and racks and racks of stuff, or building new data centers on a regular, fairly often basis, this might not be for you today or for the next year or two. Yeah, so it's interesting from like a hobby perspective, right, to see how they're boiling this stuff down. Yeah. And I think there's a lot of work to be learned from that, so I don't need to diminish it that way. But I think the end benefit directly to the enterprises is quite a way off. Yeah, I mean, I could see the other side of this, I could see this benefiting somebody like Dell who sort of said, look, I'm more in the supply chain business than I am in the technology business for servers and so forth. And I'll take advantage of sort of, in essence, free R&D or free engineering, doesn't mean I necessarily have to sell it at that scale, but they may be one of the ones from a name perspective that could take advantage of this. Yeah, and actually one other thing I put on that, when enterprises look at this and they say, gosh, I'm not going to buy an OCP server or I'm not going to build to the web scale. There was discussion about data centers in general and how do you get, you know, really low PUE with the right power and cooling and everything. And if I'm not going to take advantage of that myself, maybe I need to go to a hoster or a service provider or use services that use those technologies because they're going to be much more energy efficient and in theory should be a lot cheaper than what I could do in house too. Yeah. Yeah. Now, Greg, speaking of, Stu mentioned cumulus, cumulus is some guys we've had on the show, but your buddy, Michael Dell, made a deal with them this week where he's basically saying I'm going to take cumulus, you know, these guys who make network software, right? They're in essence, they're going after like Cisco's iOS market and he's going to put it on top of the networking hardware that he bought. So you know, it kind of begs the question like, what is Dell's strategy going to be around these types of things? They're going to be a hardware company per se, are they a hardware innovator? Are they just a supplier when they make deals like this? Like, where do you, because you covered Dell pretty, pretty, you paid attention to them quite a bit, you know, like, where are they going with these things kind of moves? Yeah, I think they're reinventing themselves, like a lot of companies are right now. If you look at IBM, for example, they're dramatically reinventing themselves. I think Dell's doing the same thing and I think that was the primary driver for them going private. And I applaud them for doing that. I think the changes they have to make, they had to do that without the scrutiny of being a publicly traded company. I think that's what probably is going to distinguish them from HP moving forward. That, you know, HP can't make those, they can't turn that ship as quite as fast now as Dell can. And I think all of this that Dell is doing is kind of, you know, obviously part of that plan that Michael has to reshape the company and refocus the company. Yeah, they're doing a lot of, they've made between what they announced this week and then what they announced at their conference. It's a lot of interesting partnerships and you sort of wonder is that they're new model where it's go after sort of cool technology through partnerships and leverage their, I guess, enterprise sales force and mid market, you know, channel based sales force and just sort of be almost, almost disrupt the supply chain and become more of an integrator than a vendor per se. I don't know. I mean, that's our cloud strategy too, right? Yeah. I mean, so it, which I think is really interesting and I don't know how it's going to work out for them. And I don't want to make any predictions about it, but I think it's an interesting model. They're doing something that other folks in the industry aren't doing. Yeah. Yeah, it kind of goes back to, I mean, Dell supply chain was always one of their advantages in the PC business and you're right, they're making a lot of interesting partnerships today. They're almost competing against some of their other businesses, right? You know, if I'm going to buy cumulus Linux on the operating system, that kind of disrupts some of the software that they bought with Force 10. There's partnerships that they have on storage that compete against some of their internal products. And, you know, Dell's got pretty thin margins. If you look at who still has X86 servers, you know, Dell's probably got the leanest margins out there. You know, HP is the other one that is trying to stay with that market. Right. Right. Yeah, and that's the question. make enough money or enough margins with giving all the debt that they now have as a a private company to get ahead of that thing, but that'll be an interesting one to watch because they're definitely taking a very different approach to the market, kind of uncharted water sort of or at least, you know, in traditional Dell models sort of disrupting the supply chain trying to take a different path. Now, somebody that you guys both cover, at least an area you guys both cover, we talked a little bit about IBM getting out of the X86 business, Cisco, sort of kind of gotten the storage business this way, didn't really, and what was your guys take on the Whiptail sort of kind of announcement and then pulling things off the wire and so forth? You know, I don't think it's that big of a deal. Well, I shouldn't say that. Trust me, I love drama in Enterprise Tech. But I don't see this. I think other folks are making a bigger deal out of this than it is. You know, from what I've read in the press about this Whiptail and Cisco is pretty committed at keeping this part of the UCS fabric. So if that's true, if that's their go-to-market strategy, and I don't know if it is or not, but assuming at face value that it is, I really don't see it being that disruptive. The only question mark I have about this whole EMC Cisco thing is VCE. There's one company that defies gravity that does things on paper, you see, you can't do that. It's VCE and they seem to skate through a lot of controversy that VCE is interesting because when they made the announcement that they said we're going to kind of align ourselves to the Cisco ACI, you know, NCMA networking, they kind of made the statement that said, Cisco is our networking company, we're going to go with that, right? More so than, say, the VMware approach to it. And they've sort of said EMCs are storage company. So it'll be interesting to see if Cisco keeps pushing Whiptail as part of the compute, then do they put VCE in a weird position because obviously EMC is going to want things like ExtremeIO and ScaleIO and various storage things, but Whiptail's technology previously was, you know, looked pretty much like ExtremeIO or Pure, so I mean, it was an all-flash array, right? Well, if we believe what they say about VCE, it's very clear that EMC is the storage piece. Yeah. Cisco's the networking piece, and VMware is the virtualization piece. Well, it's Cisco's the compute piece. And they compute piece. I'm sorry, you're right. And so I don't see them, I don't see the blurring those lines at all, but it just drives me nuts to think about each of those vendors have channel partners that compete with VCE. Yeah. And that's what defines the logic of my mind is, is how are the channel programs holding together when they're competing against the vendors that are working with it defines logic? And we've seen this, you know, whether it was EMC and Dell and that partnership or it was Cisco and HP, you know, at some point when they start sort of buying those assets or, you know, moving away from previous partnerships, like, people tend to get excited about owning an entire stack. I mean, Stu, you deal with these guys quite a bit out of them. Yeah, I'm not looking at the rinse light, but I mean, what's your, what's your read from even like when it was first announced? So yeah, I mean, I wrote it up the day it was announced. I talked to Whiptail for a couple of years, obviously, I've covered Cisco and VCE pretty intensely. You know, talking to Cisco, you know, this week, one of the things that, you know, they really were stressing that this is part of UCS. The name of the product is no longer Whiptail. It is now UCS Invicta. And by the way, if you try to call it Invicta, they're like, no, no, no, it's UCS Invicta. This is, you know, you're right, you know, it's the marketing term. This is just part of the fabric of the network of the, of the UCS managed by the UCS. And so, you know, it's all part of there. So, you know, is there a competition between Cisco and EMC? Yeah, I mean, you know, I worked at EMC for 10 years. And the first time I went out to San Jose, you know, sitting across the table from Cisco and they're like, don't you think we're competitors? And you know, a storage guys kind of looked at the network guys and were like, you know, storage is not just a protocol that runs over wire. This is a complicated thing. Storage industry, I mean, Greg, we've been, you know, storage people have been fighting for how many decades and it is so fragmented. You look at the pie chart of the vendors in storage and it is just all out war all the time in nobody is, is dominated in anything. So, you know, is, is Cisco going to step on a couple of toes and are their groups inside of Cisco that are rooting against VMware and EMC? Absolutely. But, you know, as you guys said, you know, VCE is going strong. They've got, you know, strong revenue, you know, I didn't have a chance to read the EMC earnings this week on it, but, you know, it was, it was well over a billion dollars worth a revenue that VCE is doing and they're just, you know, chugging along. So, you know, and it's not only the EMC relationship, you mean, you look at Cisco, they've got billions of dollars of networking products that they are selling and UCS especially with their partnerships with EMC, with NetApp, with HDS, with Nimble and Nimbus and TGIL and everybody else out there. So, you know, while Cisco did alienate all their server partners back when they launch UCS, I think it's going to be really hard for them to go full hog into storage despite the fact that Flash is critically important to the compute and therefore they've got to have that UCS Invicta piece to be able to build some architectures from the compute layer. Brian, I think that you're, look, people are going to call me crazy. I think at some point Cisco and EMC are going to merge. I think it's inevitable. I think it's either going to be the bloodiest divorce you've ever seen or it's going to be this Uber marriage to create the next IBM. Yeah. You know, the thing, I was looking at some numbers the other day and, you know, you look at the big six or seven or eight kind of enterprise vendors, I mean, they're all sitting there with somewhere between 10 and 40 or 50 billion dollars. So, I mean, it's not, it is crazy, but it's not completely crazy that something big could happen. Now, whether it's that or something else who knows, but there's a lot of people sitting on a lot of cash that, you know, the game is getting very interesting as they say. So, kind of staying in that space, you know, Stu, you talked about Flash, we sort of have these three big trends going on around, whether it's infrastructure or storage. You've got all Flash, you've got Hyperconverged, and you've got software to find, and you've got folks like Pure just picked up $150 million, Nutanix just picked up $100 million. You've got people freaking out about, you know, VMware potentially getting into the storage game with things like Vsand, like, which of those things, what do you see as the most disruptive? Are they all really disruptive? Is this, you know, I know Greg, you were talking to Chad about, you know, kind of the dollars and cents of it. What, what are you seeing, because you guys cover the storage space even more than I do? Which one of these is the big one to watch? So if you've limited to all Flash and Hyperconverged and software to find, Ryan, I'm not ready to say either any of those are the top ones. When you look at how much revenue the top guys are making in this, it is still very small compared to, you know, if you look Amazon Web Services, you know, estimates are they did a billion dollars last quarter, you know, that's huge. I mentioned VCE did over, you know, is a couple of billion, billion and a half last year. You know, love some of the Hyperconvergence technologies, you know, they're getting in the range of a hundred million dollars a year. The all Flash arrays are a small part of the overall Flash market. I mean, Flash is huge, if I was to take Flash is, you know, that this major wave is totally changing the way that storage architectures are going. We're five years into it and we still think there's a lot to go. I think you've, yeah, go ahead. So if we were, let's say we were jump ahead two or three years, it should cloud storage beyond this list more so than any of those or is it still, you know, a lot of disk fronted by a little bit of acceleration somewhere, I mean, like what does the storage landscape look like? I don't know, two or three years from now, even. Yeah, so is it so fragmented that you can't really pick or does it feel like it's moving in a certain direction? So I mean, we put a stake on the ground at Wikibon. We put together a new market definition for what we called server-san and this is kind of the combination of Hyperconverge, these Flash architectures and built around, you know, server-based storage, which is leveraging Flash and disk and it's software-based. So the Hyperconvergence guys definitely fit into this. You mentioned like VMware vSAN fits into this EMC scale I/O and some of the cloud type solutions would definitely fit into these architectures because it aligns more with what you see from the cloud storage solutions out there and even Amazon or SEF from Ink Tank and Gloucester from Red Hat, some of the other projects out there. So you know, we think that's huge, whether that's going to be, you know, big, big, big in the next two or three years, you know, we're probably saying more like four or five before, it's really, you know, competing strongly dollar-wise with, you know, traditional storage arrays, but, you know, we're trying to wrap up some of the market forecasts on that too. Yeah. I mean, Greg, like you guys, your company, HDS, had a big announcement about doing like object storage for one of the large service providers, I think it was, I think it was like Verizon or something. Yeah. It's the one. That's the one segment of storage, at least from sort of folks that have been in storage for a while, traditional storage that doesn't get talked about. Is that, is that kind of the dark horse? I mean, you think about all the new applications, how much, you know, how many billions of objects people like Amazon talk about? I mean, it's object. Yeah. The next big thing that people aren't really talking about that much. It absolutely is but here's I think the challenge the way I look at this is I think you're going to see more relationships and partnerships with the service providers that are at scale leveraging object storage. I think it's going to take a little bit of time for that to bleed into the enterprise, but I think there are features of object storage that are very attractive right now. And if you look across the space including HDS, I think every major vendor is going to sink and share technology that's based on object storage. But when you look at you mentioned all flash arrays, hyperconvergence, software to find, let me tease you with this. HDS is going to be making a very significant announcement on April 23rd that, well, pull together all three of those elements. So from an HDS perspective we're, I guess, the right way to phrase is very optimistic for all of those technologies. I think the interesting thing that you brought up is where does cloud fit into this and it's from a disruptive perspective. And I still think and I'm very bullish. I've not been shy about this. I'm very bullish on OpenStack. Not just because HDS is a partner into that, but I really believe that the way the ecosystems come together, I've talked with some very, very large enterprises that I think it's safe to bet. If you look at the top 50 companies in the US, every single one of them either has an OpenStack implementation. They're testing it or they're exploring it right now. And I think it's a lot of people say, let's talk about the enterprise. Let's talk about real world adoption of OpenStack. And it's there. It's real. Whether you choose to look at it or not is, you know, I think the pundits kind of selectively look at OpenStack one by an AWS another way. But the reality is every major company is looking at OpenStack right now. And that has the potential being really disruptive. When you say look, so kind of clarify that because that's the one weird thing about OpenStack is it's always, you know, people want to ask, you know, where it's being deployed, who's using it, so on and so forth. And, you know, a lot of the names, the early names, the ones that you hear tend to be technology companies. It's, you know, technology-centric groups or technology companies. So you're kind of hearing even, you know, grassroots and behind the scenes, it's, you know, it's companies like Walmart or General Motors or Nationwide Insurance. I mean, companies that you wouldn't necessarily immediately say their core business is technology technology like, you know, web-driven stuff or gaming. You're hearing that more and more? Absolutely. In fact, I know for a fact, and you named a few of those companies, I know at a very personal level, have very big budgets and are planning and architecting right now large-scale OpenStack deployments. Interesting. Are you here in the same sort of thing? Stu, are you still here in moderate? Yeah. So, you know, I'd say we've got one of those big air gaps. You know, Greg mentioned AWS. When I was at AWS Summit, it really felt like some of those early days are, remember, back to Microsoft in the '90s or VMware 10 years ago. But, you know, the deployments for Amazon today are, you know, test-ev environments and lots of new architectures. So stuff that's really designed for mobile. It's all those web-type things. It's not the old applications. And what I think I was hearing is, you know, some of the companies that are trying to, you know, really cloudify, you know, their existing environments are looking at OpenStack for, you know, part of the solution there. We just did, we did an event yesterday looking at OpenStack for the enterprise and it was still big companies like, you know, eBay and InterNap and ones that aren't your kind of, you know, non-web enterprise companies. So, you know, I think it's still early on OpenStack. I'm, you know, cautiously optimistic for it. You know, we were impressed what we saw the progress go at, you know, the OpenStack summit last year. And, you know, OpenStack, I think, is making real good progress. But, you know, Amazon is much further along and they've got a honor revenue. I mean, I just look at, you know, how many companies are adopting it. And, you know, I would be more likely to, I wouldn't put OpenStack in the same category as AWS today. I'd say, Azure has some nice solutions out there. Office 365 is doing great. Google is quite interesting, but OpenStack has not proven to me for most deployments. You know, I, there's a great ecosystem around it. Absolutely. A lot of people rooting for it, but, you know, it could make it or it could fall apart. And I'm not sure. Now, Greg, do you think a lot of those folks that you're, that you're talking to or you're here, do you think they're going to kind of come, come out public this year? I mean, do you think we'll hear about more and more of them? Or do you think they're kind of staying low because either they're just one of those companies that they don't talk about it, or they're still not sure if they want to, you know, kind of put themselves out in the limelight? No, I think, oh, well, look, let's put this way. If you're looking at a long-term project that would probably, the roadmaps I'm seeing are 12 to 18 months out to completion, what's there to talk about? What's there to announce? I guarantee you every major vendor knows about those accounts and knows the activity that's going on in those accounts. They can't talk about it either. I can't mention, I can't drop names here. Sure, sure. But I can tell you that that activity is going on, and I think every major, even at EMC, they, you know, I'm sure internally, you guys, you know where the bodies are varied too. So these implementations are real. I don't know if it's, it's not really, and they're interested to go out there way of the open stack flag, right? I mean, pick, pick any one of those companies, and which really, they're keeping their heads down. They're, they're architecting these solutions right now. They're kind of putting together these designs. I think some of them are rolling out departmental level, open stack deployments, but I don't think it's fair to compare OpenStack to AWS right now, because Stu is right. AWS is way far ahead of OpenStack. I don't think anybody would debate that. Well, where I like to do the comparison, one of the things I've been kind of pounding away at is private cloud in general versus public clouds in general. And every survey I see shows private clouds. They're both, first of all, they're both growing. So I'm not trying to throw AWS under the bus. AWS is doing great. They're growing fantastically. The, but I, an argument has to be made that private clouds absolutely growing at a faster rate than public cloud. They're both growing. It's not a zero sum game. And I think some people, I think, and Stu, I'm not referencing you specifically, I'm actually thinking of an entirely different company. There are some analysts that look at this and view this as a zero sum game that if Amazon's growing, they're doing it at the expense of the traditional infrastructure companies. And I don't think that's true. I think they're both growing. Yeah. How do you think, you know, I'm curious. So Greg, you've kind of come out in a couple of things saying you think, you know, exactly what you're saying, which is I think you're going to keep seeing the growth of private cloud. And that's going to take off Stu. I know, you know, Dave and John and you to a certain extent of, like you said, you've been fairly bullish on AWS, especially after reinvent. Like, regardless of, you know, maybe, maybe which one of those actually comes at like, how do you, how do you think what's it, what's a decent way for the industry to sort of be measuring that? And, you know, do we look at it as, you know, the pie's getting bigger. Do we look at it as it is a zero sum game and it's taking one from the other or like what, as you're thinking about those, like what sort of milestones or metrics do you use to kind of, I don't know, validate if your thinking's right? Yeah. Now that's a great question, Brian. And first of all, as you know, we've been bullish on cloud overall. It's not just, you know, public cloud. We think hybrid cloud is the way most companies are going. And, you know, it really depends on your application. So we've been covering a lot of, you know, the private cloud solutions that are going out there. So covering the shows, covering what they're doing. So you're right. It's tough to kind of decide, you know, how do we measure a metric? Because, you know, to Greg's point, private cloud is growing. But if I go to look at, you know, NetApp or EMC or Hitachi or Dell or IBM, trying to define what is a private cloud and what is just kind of a repackage version of what they had before, it's a little difficult. As opposed to public cloud, it's pretty easy for me to say, here's, well, Amazon's not quite sharing it. But, you know, soft layers, this separate product and IBM, you know, and we can go to the service providers and see what they're doing. So measuring private cloud's difficult. So, you know, I think, you know, one of the things we measure really is, are people making a migration with their applications? Because as we know, that's one of the big differences between the private cloud and the public cloud is, you know, am I, am I just taking what I have and pushing it somewhere? Or am I redesigning what I had? So, you know, there's, there's people that measure that kind of the software adoption, the refreshes there and the architectural changes. So, you know, if we see companies like Pivotal, really starting to, you know, grow and make traction, you know, I think we're going to see less in, you know, private data centers. And that's, I think, you know, a great debate is, you know, if we look out five years from now, is it going to be 20% in the public cloud or 20% in the private cloud? And, you know, it's exciting times to watch this space. >> Yeah. Are you, Greg, are you, are you kind of trying to be kind of a purist in terms of the definition of private cloud? Or how do you, like, I think that's tough. And I think, I think it's It's a little unfair to you and I referenced that there's a recent forest or report that I saw. that kind of validated the growth of private cloud, but the caveat to that was when they asked them about specific features of a private cloud, let's use self-service as an example. Well, not – and I know I'm going to get lots of blowback for this. I think if you're incorporating elements of that – the traditional definition of what a cloud is, but you're not doing all of them, yet the way you've deployed it helps you deliver IT as a service internally, and that's good enough for you to call it a private cloud. I think that's good enough. And this – it reminds me probably – I don't know, Stuart probably knows – I would say probably going back eight years ago when people were trying to measure the adoption of virtualization. And they said, you know, and people were splitting hairs, are you – well, you're not really putting mission-critical apps on VMware, you're only putting your second-tier or third-tier apps on it. And try to split hairs and make all these distinctions when at the end of the day look at what we're at in the enterprise as far as virtualization goes, are we not splitting those hairs anymore? And I think the same thing is true for private cloud. I think we – and I realize why it happens, we've got to apply some sort of criteria to the measurement so that we can be accurate in what we're talking about, when we talk about adoption. But I think it's a little unfair that if you can't check off every box off the list of how you define a cloud, then it's not a cloud, but I think it's a higher level than that. I think it's about delivering IT as a service by the definitions that are established by that enterprise, not by an analyst. Yeah. Now, I can hear that. It's sort of – it's kind of like people getting crazy about diets, and if you're not goal is to lose 10 pounds, do you have to be a full-blown P9DX paleo, or can you just run a little bit more and then lose 10 pounds, and so – well, I want to touch on one last thing, and it's sort of two things. There was an announcement today, or at least a semi-leak announcement, that Microsoft's got a new CEO. Unfortunately, I don't think it's your buddy Greg Farley, although we might – Mark Farley. Mark Farley. Mark Farley. It's Farley on my mind, and I got that common name thing on my mind, but yeah, so it's not Mark, but it's this guy Satya Nadella, which is going to take me a while to get used to, that's coming from their cloud group, which is – I've seen him speak – he's a fantastic speaker. He's obviously an incredibly intelligent guy and articulate and wealth, but he also comes from the group that loses the absolute most amount of money for Microsoft. So what does – is that more than the gaming division? Well, I think it's all under online, or – I mean, what does this – if it turns out to be true, what does it say about Microsoft? I mean, is this one of those moments where they go, you know, what the world is flipping over to online services, and we're going to get serious about it, or is this just the best guy they could find because they couldn't get the guy from Ford, and I mean, at a high level, what is – what is a higher like this say about where Microsoft's going? Who wants to take that first? Go ahead. I'll let you start. I think it's outstanding. If it's true, I think it's an outstanding move from Microsoft, and it really kind of sends the signal about this is where we're heading, is that company? Now, you know, privately, I kind of think about – and I still think about this on both sides of the coin – a lot of folks have speculated that Microsoft might be better being split up, and where they can focus on all their different competencies better as opposed to the one Microsoft strategy. And I think that's kind of interesting. I guess I'm less interested in who the CEO is, and we're interested in – are we going to see a one Microsoft strategy, or are we going to start seeing some of these different groups inside the company, start acting more independently? Right. I guess my take on this is, you know, it's kind of interesting, especially for those of us in the enterprise space that you've got a guy that knows enterprise. The big question I would have is, you know, we really haven't seen a company that's able to span between enterprise and consumer and do it well. It's just failing everywhere. So, you know, at a great point, you know, absolutely, it might make some sense to split it up, but, you know, not sure that that's going to happen. And, you know, I talked to a couple of friends of mine that really tracked Microsoft closely, and they're disappointed that Microsoft didn't go outside for this because they felt Microsoft needed some new blood. But, you know, I mean, cloud, obviously, you know, we're keeping a close eye on it. I interviewed Brad Anderson earlier this month, and, you know, he said, you know, Microsoft spent $15 billion on building their cloud. Of course, that includes things like the Skype acquisition and everything with the Shore and other acquisitions. So, you know, we'd love to see, you know, what happens here. Yeah. I mean, they're interesting. I mean, obviously, they've got more money than God. They know how to make money. They continue to make money, you know, I mean, of any of these guys other than maybe Google, they have the most cash. And like you said, Greg, if they broke themselves up, you know, some of those divisions are pretty interesting from a competitive perspective, but the other thing that'll be interesting is they're going to have to figure out, do they just completely scrap this whole plan that Balmer had, you know, this sort of, we're going to be a services in hardware or services in devices company. Yeah. They're going to go back to being a software company, you know, I think the really interesting thing here, and we're starting to leak out tonight that Gates is gone. And, you know, of course, I might love for drama in this industry. Like if Microsoft was going to get broken up, the first step of that would have to be Gates leaving. Yeah. About 10. So, you can't read too much into that. I mean, the change is going to happen eventually, but I think, you know, this could be one of many signals that we see that gives us some indication of the direction Microsoft's going to go in the future. Yeah. Could be. Let me ask you this just from an enterprise perspective, is it a good thing or a bad thing that we're seeing all these sort of major enterprise companies now also, you know, owning their own cloud? Or is it just a natural evolution? I mean, Oracle's kind of, you know, doubling down on the Oracle cloud, SAP's doubling down on the SAP cloud. Microsoft has sort of, you know, been getting their act together for, I mean, is that a good thing? Does that, is that something that just sort of becomes table stakes and everybody who wants to be a major player has to have their own cloud or what's your guys take on this? Does it just become the new silos of the enterprise? I think the more the merrier, and you know, and I think what it does is it demonstrates that there's, that AWS doesn't define the cloud, and I think that's an important message in it. And, and the fact you've got so many players in the space, and they're going to come and go, right? Some are going to drop out. Some are going to, you know, obviously you can start them where, yeah, people are going to come in and out of this game. And I think it's, I think it's good for innovation, and I think it's good for the enterprise and ultimately it's good for the end user. So let me give a slightly differing point, you know, I'm all for competition, but, you know, we really think that, you know, like the idea of some more of a community clouds, if you look at, if I can bring together people with similar data needs and be able to have access to it. So the New York Stock Exchange, you know, has been talked about a lot as to how they've got their industry data together. If I can go to some of these mega data centers like a switch or an equinex who direct connects into Amazon or, you know, others that can build data centers where I might even be having my private information that, you know, I maintain it. I choose what equipment I have, but be able to cross connect it to a large amount of data. We think that's pretty powerful. And if there's 100 big clouds out there, that's going to be tougher to do. So we would see a fewer number of large clouds being able to benefit leveraging that data, not saying that, you know, everything goes to necessarily Amazon or Google, but, you know, it's tough to move information around. And therefore, you know, we think that there should be smaller number of players ultimately, or at least some aggregators that pull some of these together. There's some of those kind of cloud consortiums out there that can help interconnect what's happening out there. And those may end up being big open stack clouds if they get their, their interoperability right together. There you go. There's interoperability in open stack. Wow. Well, come on. Randy VICE is going to listen and, you know, come on, guys, look, Randy is super intelligent. I have a lot of respect for him. And, you know, the whole API debate, I kind of was, like, completely polar opposite from him. And it has very naive about it. You hear Randy explain, you know, the benefits of an open stack API that was compatible with Amazon. Not as a means of migrating to Amazon, but how about migrating off of it. And I think that really got me opening up my opinion on this a little bit, because you see a lot of test dev work happening in AWS. And I think what the challenges that you see is how do companies that get past that test that stage, but they want to bring it in-house? How are they doing that? And I see that possibly be one of the keys to it. Who wasn't at the conference yesterday that made the comment that they wouldn't be surprised if AWS at some point adopt open stack APIs? Yeah, but you know, I wasn't sure it was Chris camp. Yeah, but that's what I thought. So yeah, and I think the fake Andy Jassy account like that one I don't know if that's true, but I mean, but that's that's the kind of dynamic that's taking place here now, right? That's that that's in terms of how people are looking at it and the interesting thing is AWS keeps their information for you know close to the best so we don't know what this what this typical or average size of those instances inside of AWS are how many real customers they actually have an in size of those customers, but there I think there's a fair My speculation that it's probably right that I'll you know a majority of that is is at a test dev level and there's something to be said about okay, if that if it's just test dev, how do we bring it into into production if we're going to bring that in house. And I think that's part of the opportunity for private cloud over stack to obviously would benefit from that. Yeah, and I mean to build on that point, right? You know, even if we're not talking test dev, you know, they're always comes this point where you say do I want to rent this or do I want to own this. So, you know, Eucalyptus has been working with Amazon for many years on this and you know, we need to be able to, you know, no matter what cloud I choose, I need to understand kind of that data portability. So that, you know, if my cloud shuts down or I change or I want to have switched to a different vendor for functionality or price or whatever, I can move that and you know, we've had this vision for a while that cloud should be much more flexible for allow me to go in and out and we're really not, you know, taking too much advantage of that yet. Yeah, yeah, long way to go. So last question, Stu. I'll ask the last question to you guys. Thanks so much for the conversation tonight. Super Bowl is coming up. My team and Greg's team are so far out of this week. I don't know. I got to know that you care. Stu, I know you're a, you're a diehard Patriots fan. Give us, give us some sense of what's it like to root for a good NFL team because Greg and I have no idea. Yeah. Yeah. So, you know, I grew up a Yankees fan. So, you know, most people outside of the greater New York area hate us. So you kind of, you know, when the late 90s, you kind of built up the, you know, era of them winning and winning and winning. And it's kind of like, well, you know, I've got enough penance sitting in my basement that it's great. And for the Patriots, you know, having won three Super Bowls and been back to another two, it's been pretty amazing. So, you know, I've been blessed to be a season ticket holder during the era of Tom Brady. But I'm still going to be rooting against Peyton Manning this Sunday. All right. Well, with that, anybody got any good Super Bowl party plans or just going to watch the game and hang out. You know, I just have one prediction. No matter what happens in the Super Bowl, the Browns still lose. I would agree with that. Yeah, great. You know, my wife's got a sweatshirt for the Browns that, you know, just in my lifetime, one please. Oh, that's right. Your wife's a Browns fan. She gets hypnotized with Greg. Well, guys with that, I'm going to wrap it up for Aaron, who's not here tonight, but thanks you guys very much for being on. Thanks for finally coming on. I shouldn't I say that as if like you guys weren't available. It's on us. Thank you very much for coming. We have to get you back on more frequently. Appreciate the, appreciate the conversation tonight. Where can folks find you guys both in your in what you do day to day and online and so forth. All right. So this is Stu Miniman. I'm at Stu. It does just STU on Twitter. Not sure if it's worth $50,000 or more. The poor end guy, but, you know, I'm a regular on Twitter and primary place to find me. My writing is wikibon.org. And I've got a personal blog. I don't write too much anymore. But if you just also Google Stuart Miniman, you'll find me because there's only one in the world. And I've got a blog on hgs.com. I've also got a video series called IO on hs.com. Actually, we've got a new set of videos that are rolling out over the next three months. The first one in a couple of weeks. And I'm also on Twitter at Nierman. And if you can figure out how to spell my last name, you win. And of course, speaking in tech, still going strong. Absolutely. Yeah, speaking in tech.com is accessible that way too. Yeah, still still being distributed by the register and still going strong. Very, very cool. Well, guys, thank you so much. Folks, if you like the show, please tell a friend or pass it along at the cloudcast.net or at the cloudcast net on Twitter or the cloudcast.net on the web. Thank you very much for everybody tonight and have a great night. [Music]