 Live from Austin, Texas, it's theCUBE. Covering Dell EMC World 2016, brought to you by Dell EMC. Now, here are your hosts, Dave Vellante and Stu Miniman. Welcome back to Austin, everybody. This is theCUBE, the worldwide leader in live tech coverage. Matt Eastwood is here. He's the Senior Vice President of IDC, runs all infrastructure and data center research. Matt, always a pleasure. Great to see you again. Thanks, Dave, thanks for having me. Furrier Sense is best. More to Husky Boys. Exactly. So, okay. What is it Dell EMC number one at now? Dave Merce. They certainly come into, I mean, from a perspective of this integration, I mean, you've got these two really strong players in storage and compute, obviously. And when you look at where the opportunities are, customers certainly are giving them a green light to do more together. So, we saw that as some of our own research ahead of the merger. Essentially, you have opportunities around convergence, you have opportunities around bringing EMC's portfolio down into mid-market and small enterprise and potentially bringing Dell's portfolio upmarket. So, those three together are pretty interesting. Matt, I'm curious. Just one of the little tactical pieces is EMC sells a lot of appliances and they leverage a lot of x86 in their components. So, how much of a bump is it for Dell as the server manufacturer to have EMC now, you know, going to change many a bunch of its portfolio over to, you know, Dell branded components? Well, I mean, the whole converged space, hyper-converged space is in transition as well. Those, all those players are in the midst of really becoming a bit more software oriented and how they think in the market. So, I do think Dell sees that as an opportunity to kind of stand, you know, kind of Dell boxes up underneath those solutions. I do think you will see, at least initially, somewhat different behaviors in the company. I think that when you look at it from the perspective of the global 2000 and the go-to-market that Billy Scannell will run, that's going to be, continue to be very prescriptive. And I think when you look at Dell commercial and where Marius will be, I think that continues to be more of a kind of choice and ecosystem play. So, I think together, they're going to try to balance this thing out and kind of have it be, you know, the best of both worlds, if you will. But kind of your conversion question, I think that will play there as well in terms of how they lead where. And actually, Matt, I wasn't speaking so much about converge, even though I totally understand that. Just, you know, things like, you know, software products from EMC, ship on a server, so. Yeah, these are impactful. We've seen it in the past. When Cisco got into the server market, it was a fair amount of angst on the HPE or HP side at the time, because Cisco was a huge OEM or a HP server. So, you could clearly see a lot of the, Dell PowerEdge being kind of an underpinning for a lot of that software. So, things like backup and recovery and so forth. But your example's a good one. I mean, you look at Chad's organization in the solutions group underneath. You got the converge guys, the solution stack, all Peter cuts, here it is. And you got Jim Gontier running. What do you want? What flavor do you like today? You name it, they have it, right? So, the observation that I've been making, Matt, I wonder if you could comment, is that Dell makes money selling anything. I mean, basically, not anything, but you know, whether it's Nutanix, I'm arguing a little bit of AWS, we can talk about that in a minute. But they've lived on 18% gross margins for decades. Right, and now this is margin heaven for them. Yeah, and just a few minutes ago when Tom Sweet was briefing the analysts, one of the first things that he said is, look, you know, kind of year one, year two of this new merge company will be very focused on cash flow. You know, they've got tranches of debt, they have some debt that's a bit more expensive than other debt, and they want to pay that down faster. So, they're going to really focus on anything they can do to drive cash, including in the lower margin space, to your point. So, you know, Dell PCs going through that enterprise sales force is an interesting play. They believe there's upside to move even more high volume product through a channel that didn't exist to them before. Dell's always been a cash flow machine, EMC's a cash flow machine. I want to ask you, Michael made a bit of a noise on Twitter when I think it was IDC data, maybe not, maybe it was Gartner data, I wonder what IDC says, of Dell EMC surpassed HPE in terms of server, unit shipments. Unit share, yeah. Now I always felt, I think I tweeted to Michael, that's nice, talk to me when you've overtaken him on revenue. What's your take on that as a market analyst? What matters more? Well, revenue clearly matters more, and I think that Dell would also agree with that as well. Michael would agree that that's how they think of them, they think of the business in terms of taking share. I think it's a milestone, I mean, certainly what it does is it disrupts the competitor story, which is we've led in this market forever. And it was just a matter of time before this happened because there's canes in the market, particularly in the hyperscale space where you have these very large deals happening. So you have two things that kind of drive that, one's hyperscale and these large Microsoft deals that tend to waffle back and forth between Dell and HPE, and the other's China. And the growth of China as a competitive element of the market, and of course HPE's strategy there is that they've divested 51% of their business to units through the H3C model. So they're not getting credit for that in our share, and I don't think Gartner shares either. Do you think that Dell has a shot at overtaking HPE in revenue, or is HPE's new focus going to allow them to maintain their revenue share? I think, of course they have a shot, but I don't think it happens. It's going to take a long time. You're looking at a company. It's a long shot. Yeah, I mean, HPE's still, they're roughly a billion dollar plus a quarter server business, and Dell's running a two and change a quarter, sorry, month, I mean, a month. You're looking at an HPE that's running at 12, 13 billion a year. Dell's probably eight, eight, nine, someone there. So they're getting closer, but they have a lot of runway in front of them before they can pass them. Matt, can you dig into the hyper scale business for us a bit? We talked about Microsoft, obviously. Satya DeLos here talked about how Azure wants to have, next year Azure stack with a Dell stack, and an HPE stack, and a Lenovo stack, and also in the cloud, they'll have those options too. So where's Dell winning, and where aren't they in the hyper scale market? So, and you've seen this from, Dell got in early with their data center solutions business eight, seven, eight years ago, and then they retooled and started focusing on kind of semi-custom tier two. I think most of the opportunity in that space is in what I would describe the tier, when I use the term tier two, I mean not your top seven or eight. There's not a lot of opportunity for branded players in that space beyond Microsoft, which is a unique kind of business considerations there. But when you go into China, or even in the US, you start looking at that kind of next tier, a service provider, that's where things get pretty interesting, and I think it does become a bit of a semi-custom model. I mean this whole space is the considerations really are degrees of integration, kind of in one axis, and degrees of customization on another, and the big guys are going to be highly customized, and they're going to do a lot of the integration and also in a very customized way. The next tier down can't afford to go quite that far, so it does tend to play a bit more to Dell's strengths. I mean IDC generally, and I think even you specifically were one of the first to sort of highlight the influence of the ODMs, this has been years ago, you guys were on this pretty early, and then Michael's response back in even 2012 was hey, we're going to compete with those guys. Are they successfully competing with them? Part one and part two is can they sell directly, for instance, to Amazon? I think Amazon's a bit more of a challenge. I do think maybe one thing I would say is we've looked at this whole build out of hyper scale cloud. A lot of it's been driven by this kind of mobile kind of mind shift that's been happening. The next phase of this is really kind of builds on the industrial piece of this, so as you start to think about what, I mean how to end the roads on just before me talking about IoT and kind of the emergence of the edge and analytics of the edge, but there's, when you start to think about where the data's coming from and the industrialization of data, it does kind of start to set you up in a conversation around workload mattering again, and the need to really optimize the infrastructure for the specific analytic workload. So it would be harder and harder for Amazon to have a single infrastructure that does everything for all workload, and you're going to start to see I think more customization around other workloads at scale, namely analytics. What about China? You talked about how HP has to kind of separate that business. We looked at the acquisition kind of merger of EMC. China was the last piece. We understand there were certain concessions there. We've seen that in the past. What's Dell, how's Dell doing in China? What do they need to do going forward? Yeah, so you know Dell's, I mean Michael's been pretty vocal that this is about jobs, it's not just about branded boxes, and that you have to look at Dell as an entity in China a bit differently than just who's game's on the box. Now having said that, you know, obviously you've got the big guys, Baidu, Alibaba, Tencent. But when you get beyond that, there's still some pretty big players in China that Dell's doing business with that are looking for expertise. And I think that expertise becomes even a little bit more interesting now that they have EMC assets in the portfolio. Because a lot of it is in that cold, warm, hot storage space and analytics on top of that. So it's really how you can bring competing storage together. So there's good opportunity for them. I think once you get past those top three players, the top three players I think are generally ODIM and Chinese branded kind of consumers. And that's really intentional to try to drive up the size and overall kind of capabilities of those Chinese players. What's IDC's take on China? Obviously you guys have a big presence there. Pat McGovern was the first, you know, publication in China, computer world. China was like 1989, I want to say. So he saw that. But China's becoming self-sufficient. They got, you know, top performance in supercomputers. They've got their own, you know, microprocessor, right? Based on open power. They want to be a global power. And they held up the Dell EMC deal a little bit, maybe slowed it down a bit. I don't know what kind of concessions, you know, they extracted. But what's your take on China as a global power? Well, they have, so there's a lot of promise there. Certainly it's a big competitive threat to all the international players, particularly U.S.-based technology companies. So they've got this growth engine of a market in China that they've been using to feed. So if you want to look competitively, you know, the Huawei's and Lenovo's sure, but also the, you know, the inspers and the suesions of the world to become fairly large players. All four of those vendors are top 10 in units in the marketplace today. And obviously the next click is how do they become more capable exporting technology outside of China? And I think, you know, they're smart and they're not going to go and try to just quickly move into the U.S., so that there's some moves there. It's going to be more focused on other emerging markets, other parts of Asia, Middle East Africa, Latin America, Eastern Europe. Those will all be places that China believes they should have a voice. Now the challenge I see for China is, and sometimes I'll compare them a bit to Japan. And I don't think that's entirely fair, but they're doing everything in a very insular way. So they've created this kind of almost closed ecosystem in China and they're using that to kind of grow capabilities. Now, if they're unique enough, then that becomes less interesting in terms of an exportable set of technologies. We saw that in Japan where they always struggled. They did things extremely well for their native market, but they didn't do things very well for domestic, for consumption outside of the domestic market. Now, China's smart enough to know that the ecosystems largely still develop here in the U.S., and they need to pay attention to that, and they need to be very fast followers. And so far, they're showing that that could work. But certainly the core hardware markets are where they're going to go really hard. And Japan's, I think there's a good comparison with Japan. There's a lot of insular comparisons. Japan's ascendancy coincided with the IBM mainframe. So my question is, how do you feel like open source will play in that equation to enable China to compete outside its own region? Yeah, some of the Chinese execs have said that to me that they're, they put people on the ground here in the U.S. and Silicon Valley to understand how these markets are developing and what it would mean to them. And then they have literally armies of engineers that they can put on it and try to chase it faster than anybody else. So they're recognizing that they can't just look into what's happening in China. They need to borrow from what's happening in the open source world in particular and bring that kind of back domestically and then use that as a way to compete internationally. So it's a really interesting one. I mean, it's a fast growing market. You know, it, years ago blew away Japan in terms of size of the infrastructure consumption in China, second to the U.S., and probably about half the size of the U.S. today. So it's going up fast. Matt, one of the things we're digging into this week is looking at the impact of Dell EMC and the channel. Just a little bit of concern we hear out there. What's the state of the channel? How much kind of stickiness with brands are there and any feedback you have as to what kind of Dell plus EMC is impacting on the channel? So that same research we did with customers, we did some channel side research as well. And what we effectively heard, and I just heard Rory Reed repeat this upstairs, is that they have a green light from both customers and partners in terms of why, they understand why the deal's happening and they see value in the deal. What they're very cautionary about is doing anything that's abrupt and potentially disruptive to the investments that they've made. So clearly the most overlaps in the mid-range storage space. What they're effectively saying is, don't do anything dramatic in terms of how, what your, my terms and conditions look like, the terms and the contracts, and don't do anything dramatic in terms of how you support the infrastructures that I bought from you. And I think that's true both on the channel side and on the end user side. And the channel, I mean, they're look, these are big bets, right? They're trying to decide who to kind of bet on, kind of for this next phase. And clearly there are always changes when something like this happens. But I think that for the most part, you have a high value channel and the EMC channel and a bit more of a volume oriented channel with Dell. I still think that they're gonna try to play that off, play those off of one another. But over time, I do think Dell becomes more of a value player. I think where things get interesting in the marketplace is over time, I think Dell will gravitate up market, more to where EMC has historically been strong, and potentially provide exposure kind of in mid-market for another competitor to come in and put to either challenge them or be an alternative source in that mid-market. We got to talk about AWS VMware. I have said that VMware capitulated, I'm on record of saying that they did. They wanted to fight bookseller comments from Carl, et cetera. But I've also made the point, back in the day, I don't know if you guys remember this, when Windows came out, Lotus refused to write one, two, three for Windows. Instead it ported to OS2 and VMS, bad call. Yeah, paying to themselves to do a quarter. Right, so smart capitulating, right? It saw the writing on the wall. Amazon's got AWS, 25% operating profit to EMC, an incredibly profitable company, 17%, 16%. So the writing was on the wall. Having said that, what's the deal mean? People are talking about winners and losers. The narrative is, customers win. True, we'd agree with that. What's your take? Well, that was my initial take was, if you're putting yourself in the customer's shoes, choice is always a good thing, and there's nothing about this deal that's bad for the customer. Now, having said that, I mean, I think your analogy or your setup is good, because if you're VMware and you want to try to, you want to try to fight and kind of fight your install base and not allow for an evolution, that's probably not the right thing to do if you look at the history of our industry, our industry has tended, if you take too strong a position and you don't give them a place to go, then they'll do something that's potentially more impactful to your business. So you could ask the question, what's really in this for VMware? I mean, if I look at it from Amazon's perspective, it's another offering that's in the store, and it's infrastructure, that they're standing up at pricing that they're going to provide to anyone for any workload anyway. So it's kind of neutral to them, but for VMware, it gives their customers a place to go that's relatively seamless, because if they hadn't provided that place to go, then they might have done something a bit more disruptive to VMware. So I think you're right, Dave, I think if you dig in and you try to resist it, I think that kind of works against you in the long run. Yeah, protecting the past from the future is never a good long-term strategy. And what I would just add to that is, VMware's got kind of this dual strategy. They've got Vic, so the integrated containers on the VMware side, and they have Photon, and the integrated side, it makes all the sense in the world because that's kind of an evolutionary and you're playing to your base, the Photon piece is still, I think, unclear what happens there, because you're trying to influence developers and you're trying to influence developers to do something that's maybe not natural to them in terms of how they would work with. But Stu, the other piece of that narrative is the Azure stack, right? Well, yeah, I guess the question, Matt, is back in the day, VMware started out on a few different types of configurations and then spread itself to support all x86. Right. And today, if I look at the x86 market, a lot of those x86's, you know, solutions are going into the cloud. So, if my overall market is kind of shrinking some, I want to make sure that my total available market for VMware can grow. So, you know, VMware's an option, even though, and I look at it, it's not quite Amazon, it's this bare metal VMware supported solution, but it helps me with that data center footprint in the cloud then. But, you know, I guess the question I want to ask, Matt, is that overall x86 trend in the data center versus, you know, the big public cloud, you know, can you just give us kind of a snapshot of where we are and where we think it's going in the next couple of years? Well, you know, it's an interesting time. So, there are a lot of bets right now on the future of the data center and x86 alternatives potentially, whether it's open power or ARM, AMD coming back into the market, you've got all these things that are kind of going on out there. Dave, you earlier had mentioned kind of ODMs and how we started looking at that. Today in the server market, about somewhere between 25 and 30% of the market, and that's in units, is what I would describe as custom today. It's either ODM direct or it's a vendor custom solution from Adele or an HPE. So, it's a pretty big number already. And that's having, so if you look at where all the growth has been happening in the market, it's really been coming out of either hyperscale or China. And that's really been the two growth engines for where compute growth has been happening and the consumption engine's been changing. And then obviously, we talked a little bit about Converged earlier, but Converged becomes kind of the play if you're trying to become, trying to protect the growth margins in that shrinking, we would call it second platform world, but basically on-prem world. So, I'm sorry, see your question beyond. But that custom stuff, that's what's going into the mega scale cloud. And then to the extent that you can replicate that on-prem, Nutanix is trying to do it. I mentioned Azure Stack before. The narrative in the press, which was fed by Amazon, was that Microsoft was a big loser over this announcement. I'm not sure I see it that way. I don't see that either. I think Azure Stack is a winner. I mean, I don't know how you feel about that. I think it's a whole in Amazon strategy, frankly. But- I think even last time I was on theCUBE, we were talking about Larry Ellison and what they were doing in the cloud. Same, same, right? And I think if you really want to net net out who Larry Ellison sees as his big competitor, it's Microsoft. And he's throwing the gauntlet down at Amazon because that's the big player in cloud. But ultimately, it's where the developer's going and who's going to be controlling the applications. Apps, absolutely. That's what matters here. And that's the bottom line, right? I mean, apps are the new control point or center point of gravity in the industry. Infrastructure is infrastructure. You got to have it. And I think, and you probably touched on this in the last session, but you have certainly the transformation of business digitally. That's having an impact on the types of applications that are being built, cloud-native, next-gen apps. And then that has an impact on the infrastructure. And it also extends that infrastructure conversation from edge to core data center to cloud. And so the conversation becomes a bit broader and a bit more complex than it's been in the past. Matt, love having you on, man. Thanks very much. Good to see you again. My pleasure, thanks for having me guys. All right, keep it right there, everybody will be back right after this short break. We're live, this is theCUBE from Austin. Right back.