 presentation. This is the wrap-up of a special broadcast of the Cube Silicon Angle Wikibon's exclusive coverage of HP's moonshot announcement. We're here to extract a signal from those. We were here all day at the announcement, following the webcast, breaking down all the analysis, talking to all the special guests, all the executives, John and Dave hitting them up with the questions, and now here to review and break down and conclude this special broadcast we have Dave Vellante of wikibon.org and Stu Miniman of wikibon.org who's in the house as well. Guys, let's quickly moderate this through. I'll moderate this guy. I want to get your opinion. As analysts, obviously we've been covering this and you guys have been we've been covering the HP moonshots at the beginning as well as other initiatives around this. I want to get your take and start with Dave. Your take of today in summary, what do you think what happened? Yeah well I think again HP is the leader in servers and they've just put down thrown on the gauntlet and said this is our new server platform. I actually got a little heat you know from people saying that this is ultimately going to eat blades and there's no question in my mind it is. HP is going to slot in virtually every type of processor into this new format and they're going to drive lower power, better packaging, greater flexibility, you know better workload tuning. This is the future of HP server platform. There's no question in my mind about it. I think they were very careful at the original HP moonshot launch because they didn't want to you know cannibalize their existing business. That's a messaging optical issue though. That's not an assumption. It was just more they don't want to focus on that but that's the reality. Oh it's true and because they have to be careful because they got to ramp this thing up and they've still got existing products to sell and you know so you know that's the balancing act that they have to play but I think the gloves are off and they're going for it and they're going to you know transition to this new environment. Now I think it's obviously not going to happen overnight. They're going to start with these specialized workloads but I'm reasonably impressed with the ecosystem partners that they have here. These guys are actually building products. They're the real deal. They have the real deal. And they let's face it we saw them on the cube today that every one of the ecosystem partners had a prop. They were showing the capabilities and you know they want to be next in line to be able to participate in this new market. So those are all good good fundamental trends. I think the question is how far into the hyperscale market can HP go. Is HP really truly you know going to compete and grab share there or is this more about trickling a little bit into the hyperscale market and bringing that capability or proxy for that capability into the traditional enterprise. And I think initially it's more the latter than it is really driving the hyperscale market. Stu Miniman of Wikibon. Obviously you know a little bit about a convergent infrastructure. You've been covering for a long time and previously in your career at EMC Office of the CTO you know what's going on under the hood down lower in the stack. What's your take on the announcement. Obviously you've been following some of the silicon that they've been partnering with and some of the software and Intel's got some other Adam announcements. What's your take and then just what's your analysis. So John I think similar to what Dave said you know HP really knows that enterprise market. They were first with the x86. And if you look at some of these these trends like software defined networking and where they're bringing these hyperscale solutions in it is tied more to the enterprise. So while most of the SDN players out there are looking for those service providers and the big web scale companies HP is trying to hit the enterprise market because that's where they live today. So the question is can they really crack some of those big accounts. We were a little surprised to see Intel as the strong partner here. And we said if you know HP really wanted to take that moonshot further and go after some of these newer applications you know maybe it should be ARM. And of course ARM's here as a partner. We expect many of those solutions to come in the second half of this year. But Intel HP so HP has a relationship with Intel going back many many generations of computer industry. Right. So maybe it's a pressure there. I understand. But if we talk about taking the you know mobile platform and bringing it to this generation Intel is not a major player. ARM is the player. Yeah. David Floyer wrote that you know Intel's in big trouble if you look at it. Don't. Yeah. Of course our CTO from from Wikibon you know it was written a post saying if you look at this trend and if you look at people like Microsoft and Apple if they port to ARM you know Intel's in trouble. So you know it's good for Intel to be here. Premier's you know partner with HP they've got a long relationship. They've got the atom they just announced their next generation ship of atom that they're pushing here. But you know if this is a true pivot point might it be something other than Intel. So exciting technologies definitely system. Some things changing in the market. So let me ask just to ask you we know this year we've been we had a martini on the cube at the end world software defined networking is very geeky. It's a lot of specific things that define what in open flow and on SD software defined networking their position. This is software defined servers. Obviously the loose definition I want to get your take on that. I mean that's not a bad thing but you know they got to be careful not to to get step in the hype factor here. They could you know there is some good stuff that we drilled that down the queue but I want to get your take on that I'll see software defined networking purists are saying whoa this is not so far to find anything. So I think that the change that we need in the industry is rather than taking general processors and trying to over optimize and have my people and software trying to make things fit the application. If you look at what Facebook is doing with open compute. They are saying we here's the kind of five to ten models and that's what we're doing. Now moonshots doing something different starting with the application it's really hyper designing those for the environment so software is leading what's happening but I'm not sure if it's software defined. There's lots of flexibility in the way moonshots build lots of different cartridges that can go in and through the software we can manage and change and optimize what's going on here. But you know I'm I'm not sold on software defined servers as as a story there but it's still early in that that's just market tech. Well so that I mean you know the fact is that nobody's really doing software defined the true hyperscales guys are doing software defined but you know he's talking software defined on the on the on the enterprise side. I mean there are some examples you know you have certain file systems that are doing scale out but in general you know none of the big whales are really driving it. I mean you've seen some some pre announcements but John what's your take on all this. On what moonshot and HP's positioning. Yeah yeah absolutely. I think it's an amazing move for HP obviously they have a lot of expertise in server. I think one you know one of the things that they were trying to do in the webcast which I kind of didn't like because this made me kind of you know a little bit but they don't tell you went back and it's like oh yeah we did the Unix servers and was kind of a historical walk down memory lane. HP's got a boatload of expertise in the server business. This is no question about it. So for them to make this shift over to this way is not groping for a market. They definitely have the chops in the server space so you know super impressed by that I think Wall Street is all is all wet on HP and I think that's why everyone was watching the webcast. I think smart money knows HP's not only well be on a turnaround relative to this area but they have a chance to pull it off so that's that's kind of my take on that. The other thing that I find really super interesting Dave is that the market place of software defined meaning the world of flash what flash memories doing in Linux kernels software developers guys who have a computer science degree whether it's low level programming down to the hardware level or just lamb stack developers these guys are powering the next generation sets of solutions from software to service to platforms of service so the cloud market the mobile market are developing huge amounts of developer expertise and those developers are the key to success in the infrastructure has been lagging behind you've seen that fumbling of the cloud and what Amazon has done is absolutely taken that market share Amazon.com went from what I used to call a junkyard for cloud Bill Jerome has now grown up with right scale and others to have a really really compelling obviously public cloud but they are aggressively incredibly getting into the enterprise to shadow IT day we've talked about this Amazon is absolutely kicking some serious but right now and I think everyone's on notice and I think that is driving a lot of change because they are the infrastructure for mobile they're also the infrastructure for a lot of you know developers so you know the issue with Amazon is people are afraid of Amazon enterprise customers are afraid of Amazon and Amazon's got competition so I think it's an exciting time I think this is a good move for HP I think it's going to be one of those things that's going to make or break the company because I think again I think it's a solid bet because of their expertise so I look at that yes so John and what one of the things we've been tracking you talked about converged infrastructure that the thing that will move converged infrastructure from evolutionary to truly revolutionary is that programmability and scale out as really the architecture that it's built on typical to converge infrastructure are not really built for scale and they're not programmable they're more of a black box solutions solutions like Natanix a little bit what IBM's trying to do with pure systems and now Moonshot do take that next step to truly make a scale out architecture that have some program you know Dave's too I think you're totally right on that one I think Wikibon has the most research out there just a little prop to you guys but Dave and I have talked to some of the top executives including a fusion IO violin and other flash guys Veriden and others and you have as well scale out open source absolutely is the game right now and I don't care what anyone says I'm completely convinced I'm seeing too many data points scale up commercial software is slowly and quickly eroding and you're going to see Oracle out there all these guys out there EMC HP IBM all the traditional vendors Dave love to get your take on this have lived a great life very prosperous life selling gear on a stand-up vertically you know scale up infrastructure with commercial software licenses VMware Pat Gelsinger he's he's not someone who's estranged this marketplace he's moving off the licensing with the hypervisor moving quickly to different value activities so so clearly scale out commodity or industry standard hardware is the game one of our guests today said you got a disk drive and then you got you know a NASA's 10x or yeah yeah 10x and then a hand is like a hundred X is an officer's math was quite right but but the the point that you're making is a good one for years server and storage vendors and networking vendors have been marking up commodity components now they've done that on the basis of hey you know those commodity components that you buy at fries aren't enterprise ready we make them enterprise ready well how do they make them enterprise ready to make it through software the problem is they've done that through not really software but through middleware that's tightly embedded and integrated with the hardware it doesn't really scale well and this whole notion of software defined in the scale out business and the hyper scale is all about commoditizing that infrastructure and again Jeff hammer Barker statement about I was just sick when I was at Facebook of putting all this money into the container and so the idea is if you can make the hardware as standard as possible and as low cost and defriiled as possible and add value through software then you can drive business value through things like algorithms and processes and apply people to create more value and that's where I think this industry is headed so this is why I come back to HP there's I think they've got an excellent opportunity here to be a proxy for hyper scale in the traditional data center because the traditional data center is just not going to all go to the cloud they're not going to disappear overnight they need companies like HP and others to provide that level of infrastructure that allows them to partially replicate what the hyper scale guys are doing I think Google and Facebook are harbingers of the future I don't think they're outliers okay Dave and Stu I want to ask final question as we wrap up the day kind of just get your thoughts together around what you think is really happening but while you're doing that I want to ask you a quick question on the front end of that is Davis to obviously HP this is not the first time I asked Jim Ganti there's not the first time they've done something compelling from a tech announcement standpoint I mean there's a HP's a lot of pretty amazing technology products that ended up into a cul-de-sac of no market obviously this market is a market opportunity with low power and high performance so the question is good market opportunity they still got competition how do they pull it off talk about the execution what does HP need to do to do this answer that question and then let's wrap up your thoughts for me but I think the first I mean they have positioned you know initially around these these edge applications and I think they got to pick those off and deliver number one number two is they've got to continue to enable other you know players I mean having Intel you know as the the the mainstay for the initial system is interesting but I think it's much more interesting if we can get some of these other ecosystem suppliers in here now we saw those guys today it's to me it's you got to start shipping those third is you've got to have the software infrastructure built up around it it takes a company like HP to actually provide you know that level of leadership and so those are really the three things that I'm looking for in terms of seeing how successful this is going to be and I think they've got a great shot at it so what do you think about the execution they have to do so Dave you mentioned some of those early applications if you look at big data if we look at the government these are relatively low margin environments and there's a lot of specialization required one of the real concerns if we look at HP is only three percent of the revenues today is done with software so they really have to build that ecosystem and rely on that ecosystem and how much of the ecosystem is going to say okay I want to go with HP so that's up their game on software yeah absolutely and this is the software defined server that we've had to see some software love the quote from the AMD guy software turns on the hardware everything's software defined so okay final packaging of this event Dave I want you to wrap this thing up for the folks summarize real quickly your thoughts impressions and what do you think is going to happen post post this announcement well John I think you know I'm honored that we were here at the original moonshot announcement outside Bill and Dave's office and the HP allowed us to come in and independently report on what's going on here at the moonshot announcement and so I think that we are seeing a major change a major shift in server infrastructure I think it's it's very consistent with the hyperscale trends we've been we've been covering this again what we cover at Wikibon and Silicon angle it's big data and it's software led infrastructure those are the two big giant trends that are being driven by cloud mobile social big data I mean everybody's sort of hopping on that but underneath that that nuance piece really is what's happening in the hyperscale with infrastructure and what's happening with software led infrastructure on top of it are really to drive new levels of value so I'm very excited to see how this thing progresses and as I say I think HP's got a really excellent shot at making it happen. Stu anything to add? No I think Dave summed it up real well it's still John as you said early innings for this technology at this point but you know big announcements expected in the second half as the partner goes from kind of one solution to lots of options and you know well still we'll be looking for your analysis obviously we know that you know you're in the lower end of the stack relative to some of the things going on conversion for structure you would put out some of the best research early on with David Floyer on this area we'll watch you. My take on kind of the wrap up here is I think what's interesting is I've sort of said my thoughts on how I feel at the announcement I think it's a home run ridiculously amazing numbers on the energy savings that alone should just tsunami of units will go out the door on that one point but the software defined piece the software led the compatibility with the future architecture of how people are going to build software I think is really going to be the key that everyone's overlooking. The other thing that I will say Dave and Stu is that I'm really surprised that Intel's got the lead position here on the announcement and I think what's exciting is HP's not picking winners here they're just probably because of their relationship with Intel and their performance with Adam but if you look at the other ecosystem partners it's a jump ball in my mind I don't think there's any decision at this point I think this product is going to be so compelling that TI arm everyone else is going to have a horse in this in this game Dave and I think there's plenty of courses so there's a lot of fertile ground for these horses to run you're going to have different tracks and different use cases of I think purpose built custom solutions I think so I like that the ecosystem's got a lot of vendors in there that are going to be built in product I think it's still going to be a jump ball. I want to add one more thing I mean the companies in this business and the enterprise business need to figure out how to make money at low cost with lower margins and that's something that HP has always done pretty well at and I think that it's going to be really interesting to see how others respond and I think software led is one way but there's a lot of talk about that I want to see more people put their money with them out this. Well that's going to be a wrap for us here for the special edition for the cube long day excellent day with HP again a lot of competition is still out there see what they can do with it I want to thank Dave Vellante for helping us out here at the cube as well as Kenny and Mick out there on the board appreciate it and all the folks back home at Silicon Angle popping out those stories Markers and Hopkins Kristen Nicole Mike Wheatley and the whole team great job and also thank HP for supporting us to come out here and be part of their great launch Silicon Angles independent coverage this event we're continuing to do more free research at wikibon.org go to wikibon.org and look for to me most seminal research on software led infrastructure we're calling it software led mainly because we believe it's not yet defined and networking servers storage they're converge they're converging fast it's software based and HP's product fits right at the wheelhouse of some things we talking about and I'll see Jeff Kelly with big data check out his new report he's leading the market and go to SiliconAngle.com is for the reference point for tech innovation covering the latest in emerging technologies and the data center and the enterprise and we're excited to be here and that's going to be a wrap for the cube thanks guys appreciate it and we'll see you next time look for silicon angle wikibon's the cube at the next event