 It's the Cube, covering HPE Big Data Conference 2016. Now, here are your hosts, Dave Vellante and Paul Gillan. Welcome back to Boston, everybody. This is our wrap. This is the end of day two, two days of wall-to-wall coverage. The Cube is the worldwide leader in live tech coverage. We like to go out to the events. We extract the signal from the noise. Paul Gillan, it's really a pleasure working with you again. It's been a while, always, Dave. So, you got little Kool-Aid injection, which is less of a Kool-Aid injection that you would normally get at an event like this. Lower key, but into your introduction, really, to the Vertica and HPE Big Data community. Vertica-8, your combinations, bringing together the whole autonomy, Vertica, open systems or open source community ethos that they've done, that product of Haven. Starting to get some market traction, we heard from a number of customers that were using it, some paying, some not. But so, you see HPE, we had Robert Young-Johns on. He told us back in June that Discover that he had trimmed some of the fat, they sold off Tipping Point to Trend Micro, rationalizing some of the organization, all along, all the while, Vertica, chugging along, sort of doing pretty well. Swirling rumors about private equity, really, people aren't sure what's going to happen to this division, but nonetheless, this tight-knit group within Vertica seems to be doing pretty well. Seems to be almost an enclave in and of itself, isn't it? I mean, we didn't hear, we heard nothing about HPE hardware during this conversation. Oh yeah, they heard about HPE storage. I mean, this was about software, it was about analytics, about database, and it was almost like HPE was just an addendum. Well, and that underscores the challenge and I think the opportunity. I mean, I always felt that, I've said for years, HPE's got to do, HPE now does do better in software, has to have a larger contribution from software, that's got to be a fundamental part of its strategy. I'm not sure it is. I'm not sure that Meg's goal is to be relevant in the software business anymore. It certainly was back when the board, way back when under Apatek and Bot Autonomy, they saw that as an opportunity, Mark Andreessen sits on the board and was saying software was going to eat the world and so they took that to heart. It seems like HPE is comfortable being an arms dealer to the cloud and duking it out with Dell as the alternative to cloud, public cloud. And you know what, that kind of duopoly is not necessarily a bad business. But it's not a duopoly, it's only a duopoly on these shores. I mean, you've got all of the Chinese competitors, you've got, I mean hardware just seems to me to be a race to the bottom. And IBM got out of that business, I think that was a smart move. I question why HPE would want to double down on the hardware business when it's clear that software is where the growth is going to be, where the margins are. It's, I mean, they've got a diamond in the rough here with Vertica, clearly a product that the customers love, a product that I think they did a good job at this conference, at least for a newcomer like me, of defining where Vertica sits in the stack. You know, it's not a competitor to Hadoop, it's a complimentary to Hadoop, it's not analytics on production data like Microsoft and SAP are doing, it's a dedicated analytics engine. They did a good job of defining Vertica, the strategy for Vertica as being go where the data is and I think there's going to be, we're going to hear Internet of Things kind of evolutions coming out of that as they, look what GE is doing, a series that theCUBE just did with GE on their predicts project. GE is very much focused on getting analytics close to the source of the data because when you have to detect a failing machine component on the factory floor, you can't afford to send that data back to the cloud and analyze it, you've got to make that decision there. And I was hearing the same kind of message here from HPE and that's going to be powerful as the Internet of Things comes online. So I want to pick up on that theme, hardware race to the bottom, why would HPE want to be in the hardware business? And it's interesting, brings up an interesting conversation. I think the answer is, well they're good at it, okay? They've learned how to make money in a lower margin environment, but very clearly margins are getting depressed in this business, EMC couldn't go it alone, right? It was too much pressure from Wall Street, a lot of pressure on margins, a lot of pressure on stock buybacks, funding R&D and maintaining 60% gross margins, high 50s, pushing down. And so very, very difficult business, so you've got to have a lower cost structure, that's clear. You're bringing up a point on China, which is really interesting. Now I will say this, ever since I've been in this business, which is now becoming a long time, people have been saying the hardware business is dead. We saw it with the mainframe, we saw it with open systems, we're seeing it now again with cloud. We'll see, China is a real wild card, there's no doubt about it, and you're bringing up an excellent point. China wants to be self-sufficient, it's got, I don't know, some number four out of the 10 top supercomputers, from a performance standpoint, might even be higher than that. By the end of the decade, it's going to be independent in terms of being able to fuel microprocessors without Intel. It's got its own version of a Linux operating system, and the Chinese government is clearly trying to buy time for Huawei and Lenovo and the likes of others, and so that's a really interesting dynamic and a scary one. Does the whole hardware business go to the way of the disk drive business, right? It appears- I think the trends, certainly if HPE can find a way to make hardware into a thriving growing business, they certainly will be bucking the conventional wisdom. Well, and Oracle's trying to buck the conventional wisdom by building the sort of iPhone for the air-prize. Yeah, and they have their own data machine, and certainly, Nutanix, I mean, there are new plays in hardware that are exciting and growing. Well, but even Nutanix, I mean, okay, yeah, appliance and all, but it's all about the software in that case, software defined, and so HPE, potentially, not potentially, it is basically saying, we are going to be an arms dealer to the cloud, high volume, lower margin supplier, exact opposite of what IBM's doing. IBM's getting out of hardware. You said it was a good move. It was a good move for IBM, but IBM doesn't like hardware. Hardware's kind of a schlocky business. I didn't used to like hardware a lot. They're not really a schlocky company, but hardware has changed. And so it's going to be really interesting to see where this whole thing goes and what HPE does with its software division. Needs to provide clarity. Hopefully at this earnings call, we'll get some visibility on that and really understand if all these rumors are true or if they can put them to bed. It's likely that they're exploring these strategic options. As you pointed out yesterday, they have to. Well, they have their earnings announcement, I believe, coming next week. And they're certainly going to be asked about that on the earnings call. And they're going to have to, they have to do something about these rumors of taking a private, taking the company private. It's, I'm sure it's a huge distraction inside HPE right now for the employees. So they've got to make some sort of statement that guides, that directs people to where they're looking. And if the company goes private and private equity, you have to really wonder what the hardware, what's going to happen in the hardware division. Because the whole history of private equity in this field has been to get rid of hardware. Private investors don't like hardware because they see it as a race to the bottom. I don't know if that's true. I mean, Seagate went private equity, and then gave, so there's been money to made. You saw Dell go private equity. I actually, you know, we'll see what happens there, but I actually think Dell's going to make a lot of money on this deal. Again, lots to be seen. But the interesting thing to me is the trend in private equity with companies like, you know, BMC, CLIC is now going private. Obviously Dell, EMC now, private, potentially HPE, or at least maybe a part of HPE. Rackspace. Rackspace, private equity, I think, sees these undervalued tech companies as an opportunity to make money in the longer term. But to do that, there's got to be investment. It's not just going to be suck the cash out and collapse the company. It's going to be invest, reposition, sell off some assets, refold it, maybe float it again. I think we're seeing sort of a new breed of private equity investor come online now that is actually looking at companies, acquiring these companies as an opportunity as they are trying to pivot their business models away from software licenses and to more of a cloud subscription model. They can get some breathing room by going private for. In for a great example. Yeah, they go private for three or four years, make that pivot out of public view and then perhaps come out again. I do want to, you brought up IBM and we both brought up IBM. One very in your face statement that Robert Young-John made today was about taking on Watson directly. They see the Haven APIs as being a superior alternative to Watson. They see IBM as taking Watson very much in vertical industry directions, ad tech, retail, healthcare, which is not a bad strategy for Watson, but they think that HPE can focus more on those horizontal segments such as manufacturing and marketing. And he's inviting developers to try them both. Use the Watson APIs, use the Haven APIs, and he says we'll beat them and that's pretty in your face. Well, so we're doing that. So I was texting this morning when I saw that video, we got our developers and our product team to start playing around with it. The results aren't there yet, but we're doing a little bake off with just one of the APIs, the video translation, transcription, which is of course a big business for us. We do a lot of video. We use Google today. We're going to test and see how the Haven API stacks up against it and make sure we're using it properly to see. There's a lot of robust capabilities we're told within the Haven API. There's 70 of them more than what's in Watson. Now that in and of itself doesn't mean much, but and we've played around with the Watson APIs a couple of years ago, and they were really immature, more recently playing around more, they're getting better, still not quite there yet. So it's exciting to see the marketing videos, but when you put the APIs in the hands of the practitioners is when you see all the warts. So I think we still have some ways to go there, but I come back to HPE's software business, and I've been asking the question or been making the statement HPE's got to do more in software. What has it got to do to be more relevant in software? Security, obviously a good business for them. Vertica, obviously a good business for them. It's just sort of piece parts that Robert Young Johns is trying to bring together. I think you're right. Vertica is a diamond in the rough. No matter what happens, that is a winner. We talked yesterday about Vertica sort of being in between the enterprise data warehouse and the big sucking sound of Hadoop and a viable platform to make money in big data. Yeah, I think your point yesterday about Hadoop, about the big sucking sound is an opportunity for HPE where clearly Hadoop is sort of entering the trough and there's a lot of disappointment now with the shortcomings of that platform, not necessarily from a technical perspective but just an implementation perspective. And HPE has an alternative or a compliment depending on how you see it to Hadoop that is clearly, that is a world class that is battle tested, is supported by a respected and reputable company and they have an opportunity here perhaps to take some of those Hadoop implementations that are floundering now over the complexity issue and bring them into the light with a really robust analytics engine. One of the things that HPE, if they're going to bring the software business more to the fore, needs to do is talk about it more and we don't hear HPE executives talking a lot about the company as a software company and that may be just my own perception, but- Well, they used to talk about how they were six largest in the world and they had aspirations and they've toned that down probably for obvious reasons. But regardless, young John's kind of laid it out. He said, developers are driving infrastructure decisions. The world is going sass, no question about it. Line of businesses are making decisions, the whole shadow IT thing. Open source is a major factor, you have to contend with it. Containers and microservices are a new trend to allow portability, you've got to plug into that and their big bet is on analytics and that to me is how HPE becomes relevant in this whole software discussion is it focuses on analytics, keeps focusing on R&D, it offers choice, it leverages autonomy. Yeah, okay, it overpaid for autonomy, but there's tech there. Well, maybe- There's real tech there, right? Maybe figure out how to make that work. Massive global distribution channel and partnerships like the one with Microsoft. So take the young John's playbook and the HPE strengths to the extent that it remains part of HPE and you could have a winning formula. Another thing I just want to point out, Dave, that young John said was the days of a huge data store are gone and which is kind of an anti-big data statement but I think it's actually a pro-big data statement because what he's saying is that data, we need to save everything. He said that on the stage this morning. We need to go into the future expecting that we will save all of our data. Well, you can't do that in one central data store. So you've got to figure out a way to work with that data when it's at the edge, when it's in the department, some of it's in the cloud, some of it's on-premise. That's going to be a big architectural challenge. And that's just shot at Oracle, obviously, right? So, okay, and that's a whole nother dynamic. We don't have time to talk about it. All right, I think we've got a wrap here. Again, great working with your two days of awesome content. Our colleagues out in California are still going live. Check that out on siliconangle.tv, Stu Miniman, John Furrier and the whole team out there. Alex, Patrick, Greg, Miriam. Thanks very much, you guys. Awesome team. Awesome team. John, fantastic. Okay, so we are, let's see, what's next for us? The fall season's coming up. We've got many, many shows. We've got a show at the Rosewood next week, right? We've got Riverbed coming up, and we've got Big Data NYC as part of... Speaking of going Prada. Yeah, another one, Riverbed, right? Coming out of the woodworks. IBM Edge is coming up, Oracle Open World is coming up. These are all shows that theCUBE's going to be at, and then some. So, check it out. Check out siliconangle.tv for our upcoming schedule. Check out wikibon.com for all the research. And siliconangle.com, you'll see Paul's work and his team and Rob Hofe and Kristen Nicole and others who have been live blogging this stuff you know, daily, so thanks for all your efforts and thanks for watching. Thanks to our community. Thank you to our sponsor, HP. We're out, this is theCUBE. We'll see you next time. Bye.