>> [Voiceover] Live from Silicon Valley, it's theCube, covering Mobile World Congress 2017. Brought to you by Intel. >> Okay, welcome back everyone. We're live here in Palo Alto, California inside theCube's new studios, 4500 square feet in Palo Alto, just opened up last month and excited to be here. Breaking down Mobile World Congress all day, from 8 a.m. to 6, today and tomorrow. As their day ends, we're going to pick up the coverage, do the analysis, get some commentary and reaction to all the news and also the big trends and my next guest here in the studio is Saar Gillai, friend of the Cube, Cube alumni, and former HPE Senior Vice President GM of the teleco business. Ran the cloud, then a variety of things with Meg Whitman at HPE, now he's independent board member and in between gigs on the beach, clipping coupons as we say, Saar, great to see you, looking good. >> Great to be here, nice studio. >> I'm excited that you could come in, this is exactly why we're having our show here in this new studio because a lot of folks that don't take the big trek to Barcelona who don't have to can come in and talk to us and you've been a veteran of Mobile World Congress for many years. Again, you ran, and actually built the cloud biz and also built the, I won't say NFV biz, but essentially the teleco communications division for HPE so you know a lot about what's happening in the industry and more importantly Mobile World Congress. This is the year that all the accelerant is coming to the table, all the rocket fuel is being poured out on to the bonfire, the matches are going to be lit, it's called 5G, it's called IOT, internet of things, internet of people, the devices look good, they all want to be Apple, they all want to be over the top, running entertainment, smart cities, cars, 5G is the holy grail, we're done. But seriously, where is the meat on the bone on this thing? Is it real, is this transformation hype or reality at Mobile World Congress? >> Yes. (laughing) >> Yes, hype or yes, it's real? >> There's a lot of hype, but there's some reality. I mean I think first of all, 5G is the latest thing, it used to be LTE, now it's 5G. What does 5G actually mean? Really, for people, what 5G means is you should have a lot more capacity, right? So 5G talks about even up to one gigabit in certain cases, lower latency and so forth. Now the thing about wireless is you know, there's no secrets in wireless, okay? It's not like... >> It's a physics game. >> ...Yeah, it's physics, it's not like ethernet, where you can go from one meg to 10 meg then all you have to do is run more line and you're good. If that was so easy in wireless right now, we'd all be getting one gigabyte right, but we're not. So the only way you increase capacity in wireless is through smaller cells, and there are some mimo technologies and so forth. You know, 5G talks abouts technologies that will enable you to do that, but it's much more of an evolution than revolution and people need to understand that. There's no fundamental shift, what they're talking about in 5G is adding a lot more bandwidth. Today, most of the frequencies being used are sub five gigahertz, those are great frequencies to go through walls, they're not that great in terms of capacity and there's not that much of them. Like AT&T might have 60 megahertz, that's the entire capacity they have in the U.S. And that's not much. And so they're talking about using millimeter waves, other things like 27 gig, 28 gig, 60 gig. Now, those do have a lot more capacity, they have other problems, they don't go through walls. So, I think instead of thinking about 5G, we need to think about, okay, what problems are we trying to solve? Like, what problems is this going to solve? I think in some sense, it's... ...while everyone wants more broadband, some of it is a solution looking for a problem. >> [John] Yeah, it's a field of dreams too dynamic, build it, they will come. That has been a network operator concept, right? And then we know the operators, and you and I have talked about this on theCube many years, the operators are having business model challenges, problems, challenges, there are opportunities, but at the same time there is a bigger picture I want to get your thoughts on. So in a vacuum, there's limitations, there's physics, but now, you're looking at a connected network, and this is the end to end concept, so under the covers of wireless, assuming wireless has its topology, architectural things you could do, smaller cells, different frequencies, how it's going through walls is preferred, longer distance, longer latency through walls, that's the ideal scenario. But I think there's a bigger picture around the different types of wireless networks, but there's cars, there's mobility, actual true mobility, 60 miles an hour in a car versus walking down the street or sitting in a stadium or at home. These are use cases. How much of it is a wireless problem versus another problem? NFV, end to end, virtualization, help us parse that through, how should we think about this? >> There's two issues, there's a wireless problem, we can talk about the different segments that make sense and don't make sense or how much they have to evolve to make sense. And then fundamentally, the networks are very... ...they're not that agile as we know. Which is why NFV really, if you remove NFV, and you just... ...NFV's about creating agility, and yes their doing for virtualization and yada yada, but it's about creating agility, creating automation, right? You can't have these... ...a lot of these networks were designed years ago even 3GPP, this is a decade old. And so yes, there's a lot of work that has to be done and creating much more agility in a network because the network isn't built for that. Just if you think about even simple things like number of subscribers that can go on and off, right? Okay, if you have a cell phone. Like today, if you look in the world there might be 80 billion subscribers, lets say. If you look at the number of cell phones and so forth. But once you start IOT, you might have 100 billion because every device will be a network. That's a different management system, right? Also, those devices may go on and off every day, right? Because you buy a new device, you plug it in the wall. Okay, phones, you don't start a new phone every day, right? People buy a phone to use it, so, the network becomes much more dynamic, the back end has to be more dynamic, that has a side effect and so there's a lot of work that has to be done on the backend to make it more dynamic. That's the backend problem and then, you know, they are working on it. >> [John] And the bright spots there are what? What are the bright spots happening today, this week at Mobile World Congress and the trends around the backend? >> Well, you know, I mean Mobile World Congress is a show, right? And these are not sexy things, so we probably won't hear a lot about them, but you hear about orchestration, automation, network virtualization, basically moving all this through the cloud paradigm, where you have a lot more flexibility. I mean if you think about what's happening NFV these days for example, you don't hear a lot about it, but what is happening a lot is onboarding work. Okay, we've talked a lot about it, from that hype, now we're into build-out, right? So you hear less about it, but stuff is actually happening. >> [John] So it's operational. >> [Saar] It's operational stuff. >> Yeah >> [Saar] They're modifying the system so that they can be ready to work when you get to that point. On the radio side, I think the important thing to understand is like you said exactly, there's multiple use cases for 5G. The most interesting and immediate one, potentially is to use wireless to compete against cable. Which is fixed wireless access. You know, there, the telecos for years have wanted to do it, there was this whole discussion about fiber, then it turned out fiber's expensive. >> [John] Yeah, you've got to trench it, you've got to provision it to your home, you've got to roll a truck. >> Took Google a few years to figure that out, but even for Google, it's expensive. >> [John] People who have done that said you're crazy, but Google's has had so many deep pockets and Facebook does the same thing with their kind of R&D projects. >> They figured it out, there are technologies, millimeter wave is a bit hard because it doesn't go through walls, but I think when we talk about capacities, it's not for your mobile phone, it's for other things, it's mostly for fixed wireless access. There's a whole discussion about cars, I personally, because we're talking about opinions here, I don't understand the problem so much because the reality is the car's going to be a mobile data center. So 90 percent of the data that's generated by the car will be kept in the car and the car will be sending analytics and metrics up so it doesn't need gigabits. It's not like every time you turn you need to get it an instruction. Maybe that's what the network guys want you to believe because then you need like zero latency. But you don't need that, it's much easier to invest in a better system in the car. So the car's not going to figure out... >> [John] The car is a computer, it's not a peripheral. >> Yeah, it's a peripheral. >> [John] It's a data center to your point. >> [Saar] It's a full data center. It's the edge that computes, so I don't think that's an issue, I think the car will need coverage and so forth. >> But that's a different thing, cars are great examples so let's take on this one. 'Cause this is a perfect mental model. A car is going to have all this capacity like a big computer >> [Saar] It's a data center. >> [John] Or a data center. >> [Saar] A mini data center. >> A lot of things, a lot of instrumentation, a lot of software, glue. >> [Saar] It's going to have 10 computers, big systems, it's like a little data center. >> But it also moves fast, so it's a true mobile data center. So it needs mobility. So mobility has trade-offs, right, with the wireless piece at least. Depends on how you're uploading. >> Again, it depends on the capacity. Mobility has certain elements when you get into Doppler Effect and so on. It's always a capacity trade off. All of you have talked on your cell phone or used data on your cell phone in your plane, we know this. When the plane's landing, that's 150 miles an hour when the plane lands, okay, and it works pretty well. >> We all cheat, don't turn on your cell phones, we're landing. >> [Saar] Yeah, exactly. >> We've all done it. >> So again, if you want to run a gigabit, it's a problem, if you want to run less it's not such a big deal if a car's going 60 miles an hour. So it depends. Now if you define the use cases, I need a gigabit for every car, there's a million cars, that's a problem. If you define the use case, as something else it's not a big problem. There is a problem though, and I think that is something that the 5G is trying to address in terms of more on the backend of density. >> Density in terms of signal, or density in terms of... >> In terms of support, so for example, the one place you can never use a cell phone is in a conference, because too many people are trying to get on at the same time. It's a statistical model. >> [John] A-station issues. >> It breaks, and so with a car also, you're going to have high density because if you have a traffic jam, all these cars are talking or receiving, so that's a bigger issue. And 5G does talk about that as well, but that's a bigger issue than pure capacity. Pure capacity, great, I'll give you this much megahertz... >> [John] I agree with that. >> You can do that on WiFi today. >> [John] I totally agree with that. So let's take a step back I want to get a little color on Mobile World Congress. Talk about what's going on right now. So it's dinner, people at parties, what goes on? People want to always ask me, John what always happens? First of all, Barcelona is a great city as you know, we've been there together for some HP events, as well as for Mobile World Congress. What's happening, you always make the comment it's a Biz Dev show which means it's business development going on. All the top executives go there, deals are being cut, but it's also a large trade show as you will for mobility. >> I think like you said, from my experience, the biggest value of Mobile World Congress is not the show itself, with all due respect to the show, it's the fact that everybody and anybody who is somebody is there, that's why we're not there. So you can meet people. And so if you want to meet a bunch of people, teleco leaders and so forth, that's what you do. This is the place. You all say, okay, we'll meet at Mobile World Congress. So like for example, when I was down there I would basically go back to back from in the morning until 10 at night, in meetings, dinners, whatever with CEOs of various telecos or CEOs of partners and so on. Everybody's there, and I never actually got to see the show because I never got out of a meeting. And most of what happens there is that. That's amazing because again, everybody's there. >> [John] There's a huge ecosystem involved. Talk about that ecosystem because this is the dynamic. And first of all, we don't have to go there because we've got theCube here so we're there virtually, digitally, and that's what we do now. This is great, in the studio, we save ourselves the three day flight to go to Barcelona. It is crazy there, but it is about the community there, because you have that opportunity to get the feedback, do deals. >> [Saar] A lot of deals around there. >> [John] A lot of deals happening, also feedback, trying to connect the dots and having the right product strategies. What are some of the things that you think is happening right now from a business standpoint in these meetings, right now? Are people still scratching their heads on over the top, is it the classic problems, what's the current state of the union? >> Well, you saw Vimpel Com change their name to some other thing, so I think what you're seeing right now, is there's still sort of multiple dynamics going on. One dynamic is there's people maneuvering around how 5G ends up closing and there was some discussions about that, there was some release done about hey we should speed it up and then Enrique said no this is silly. So there are some discussions, there is some maneuvering going on like any time when you're doing a spec, when does it freeze, when does it not freeze? Some of the telecos want this, and so forth. That's sort of in the background going on. They're still trying to figure out, you know, business model is still an issue. The people are experimenting and you're going to see a lot of that, experimenting with apps, experimenting with these monetization strategies. So there's a lot of that going on, trying to figure out, okay, how do we monetize the network in a better fashion? >> What do you think the best path is from your perspective? Just putting your industry hat on, if you had to kind of lay down some epic commentary to the teleco bosses, hey you got to cannibalize your own, get out in front, what would you advise them in terms of what to get out in front of, what to double down on? >> I think some of them are actually doing this, but I think first of all, I think they should forget about worrying about the technology. I mean, technology is very important, we need to take care of that, but really, they need to know what are they good at? What are they strong at? So their strong at a customer relationship, they have customers that they quote unquote have as partners, those customers, and they're very strong so what can you do with that partnership as opposed to all kinds of other random stuff. Now, if you look at what they're doing, they're doing different things. Some of them are like buying different media companies, so there's no easy path, but they're going to have to use their strength as opposed to try to become somebody they're not. They're not going to become Google, they're not going to become Amazon, they're not going to become one of those guys. They do need to become more cloudified just to be efficient, but that's because that's sort of the... ...just to play, you have to pay that card, but they're not going to be better than the existing, but they do have a very strong relationship with customers, they could probably sell them more things if they focus on good customer service. Customers are happy to work with them if they get a good deal and a frictionless environment. So, I would certainly encourage all of them, and I know many of them are focused on this, to improve your frictionless interaction for the customer. If the customer has a frictionless interaction and gets a good deal, they'll do business with them. >> Are you worried about the teleco's customer relationship when they have this decoupling trend kind of happening where the consumers want to take their phone or device and uncouple it from the network and just add more mobility across networks. So if there's better connectivity, I could be able to hop between Verizon, AT&T, whoever, that seems to be something that a lot of folks technically are saying from an architectural standpoint, having that personal centric view versus a network centric tie-in. Is that on the radar at all? Or is that still kind of in a way, fantasy? >> It's like people are still using AOL, right? >> [John] Who? Who? (laughing) >> They voted for Trump then, huh? >> I'm not going there right now, we can discuss that later. The point is, the primary area where there's problems in that area is roaming. And there's a lot of discussion about roaming. 55 or 60 percent of people turn off data when they go overseas because the roaming fares are so incredibly expensive which makes no sense. Why would I have a longer cost because I happen to have an AT&T contract in Europe, I'm not using more data than somebody in Europe and it's going through the backend of the internet anyways. So I think there... >> [John] It's a great way to jack the user with more fees. >> But that's not sustainable. >> Of course. >> I think there, you're going to see pressure of people and there's some companies who provide apps, and cards and sim cards, but there's now soft ways of doing it, there you're going to see pressure and I think eventually that will go the way of the messaging, where... >> [John] Like WhatsApp >> Yeah, they'll come up with something that will allow you to have data at a much cheaper rate, I don't know, does it make sense to switch carriers in the local market if you have a good price? I mean, what's the point? So again it all comes back to, do they give you a frictionless service? If they give you a frictionless service that is at reasonable cost, then you'll use it. So you've got to look at places... ...where their going to have people leave them is where they don't do that. And there are places they don't do that, roaming is one of those places. >> [John] So I've got to ask you about IOT, obviously it's the hottest trend, AI's more of the mental model that people get their arms around, they see virtual reality, augmented reality, they call that AI, it's more of a mental model, it's really not AI, but IOT is really where the action is. People see networks, where devices as you mention are coming on and off, you just don't provision those as static devices. They're very dynamic. Your take on the IOT market, what's your view on that? Because a lot of action happening there. >> I've been involved in IOT and different people have different names for what T means, I won't go there here. >> [John] T and P, things and people. Let it be watch... >> [Saar] Well T could mean things, it could mean other things too, but the point is, IOT ...I was working in a company that was doing IOT when we called machine to machine, if we had called it IOT, it would have been better. The point is IOT is, this is extremely fragmented, it's a super super fragmented market. And it has different ecosystems. The more complex part of IOT is not the front end, it's the backend of how do you manage devices how do you tie them to some app, how do you configure them, provision them? >> [John] 'Cause of the backend, infrastructures are different. Some are IT based... >> Think about it, you've got all these devices how do you upgrade them? How do you make sure they don't start a denial of service attack on their own? How do you provision them, how do you manage their life cycle? HPE has some product in that area, a global connectivity platform, but other people as well. So, this is a bigger problem. The backend is a much bigger problem than the front end. What's the problem? Hypothetically, I can stick a SIM card into anything and it's now a device. Most of these things do not have a high bandwidth. Low bandwidth coverage is pretty good in urban centers, not if you go to Utah, but other places. So, the biggest problem is backend. Now obviously there's a lot of advancements that can be done on the front end too, because of power issues. The biggest problem with IOT is depending what you want, you have a power issue. For example, we used to do this back in the day, you built these little devices, you stick them on containers, and then you can find out where the container is at any given moment. That's great, but how long does this thing last? I think IOT is a very big thing that's happening, I think most of the problem in IOT is not in the front end, it's in the back. >> [John] Yeah, I would agree with that. Also it allows you to get more data too. Another problem is storing up more data which is security, data, IT management, basic stuff. >> [Saar] Very basic stuff, and that stuff is hard to fix because again usually IOT is not a green field that are going to connect to something that exists. You're just augmenting it with IOT like if it's a power meter or something so now you have this existing ecosystem that has to interact with something that is brand new and so there are various companies who build interfaces and how to solve it, there's management issues. But I think IOT is real. >> [John] So let's talk about cloud. So cloud you also had your hand in at HP as well, you had a wireless background, the folks might not know that, going back before then. The cloud really is an opportunity, we see that with Amazon and then Microsoft's now got their stock up and so obviously cloud, it's a bigger game, it's hybrid, it's happening and then you have all these other fringe things developing around the mobility piece. How is the cloud changing the Mobile World Congress game? Because now it's a show that kind of blends. It feels like CES on one hand, it feels like Cloud World on another. It feels like IOT and teleco world, and all these things are kind of in a melting pot. >> Well I think to me, when I look at Mobile World Congress, I think of okay, it's teleco world, really because whatever the telecos happen to be doing is what the show is about, right? If you think about the telecos, we're talking about companies that have a capital like I think AT&T spends like 20 billion dollars a year or something in that range. We're talking about hundreds of billions of dollars of capital budget and whatever those folks are interested in is what shows up in that show. >> [John] So it's still a Teleco show dominant, you don't see that changing at all? >> No, but what teleco is, is changing, right? I mean they have broadcast, so what I'm saying is whatever the telecos are interested in is what shows up in that show. Drones, cars, telecos have their hands in all these things. And that's why it shows up in the show. Because ultimately the show is about the telecom space and the primary players. If you go down there, the booths that are as big as a country club, you know the Ericssons and so forth, it's a teleco show and people who want to be relevant like Intel as they want to be more relevant in wireless is building a bigger and bigger booth over there. But what teleco is really is about connecting things and as there's more things to connect, telecos get involved in other things. If they see a business opportunity, for example drones, lets go back to talk about that because there's a whole drone day and all this other stuff there. I mean drones need higher... ...I don't think they need so much bandwidth although it depends what kind of video you want to do. But they do need reliable connectivity. It's something useful, right? And today, you could argue connectivity it's not super reliable, it's pretty reliable, but we all have dropped calls every five minutes, right? I mean if a drone drops a call, that may not be that easy. There's use cases around these things >> [John] And back to your earlier point, I think this is the most important for the folks to listen to and hone in on is that there's a use case in every corner depending on the view of the market. Drones is one. Take virtual reality, augmented reality, that's another. IT, enterprises connecting, entertainment over the top, smart cities, these are all kind of nuanced areas. >> But when people want to understand, separate the hype from the non-hype, see if you can understand the use case. If you can't understand the use case or if the use case seems out there, then the technology is probably out there. Technology on it's own is fascinating, but if there's no use case that makes sense right here right now like again, for example, if I got a gigabit to my phone right now, would it make a difference in my life? An extra 20 hours of battery life would make a difference in my life more than a gigabit. >> That's a good point, right now battery life is more important than connectivity, but as the network transformation which is a big buzz word for this show is coming to the surface, that's an end to end architecture with software. So we think about traversing cloud, software, delivering of apps and services that's different. Now the apps have more headroom in that case, but then to your point, the backend's got to be... ...under the hood has to be smarter. Or is network transformation not yet there? >> Well I think what's happened is that the OTT, the OTTs started developing 20 years later, and surprise surprise, when you develop 20 years later, you have advantages, doesn't matter who you are, so their backend is a much further generation than the teleco's backend and so that's why when you connect to OTT services, it's consumer experience, it feels seamless and so forth, when you connect to the telecos backend, it's sort of a mismatch. And so they need to sort of fix that and that's part of the NFV transformation they're working on and again it's not because they had any limitations its because they had existing stuff, it's much easier to build from scratch. >> Final comment on Mobile World Congress this year, and outlook for the next year, your thoughts? >> We need to parse out what actually comes out of there. It's still early, I think 5G, 5Gs going to be what people are going to talk about, this is the thing. It means multiple things, but that's because the entire teleco world, if you think about it, if you look at the revenue of the suppliers and so forth who have been in a holding pattern ever since 4G sort of, in China, they finished 4G deployment and so the next big capital spending is going to be 5G, and so you're going to see the providers push anything to get that going. That's just the bottom line. >> Great. Saar Gallai, final comment, what are you working on now? Obviously we got to know you at a personal level with HPE, I've seen your roles, and the last one was really handling that teleco business which you grew up from a handful of people to hundreds of people, thousands of people. You're a land grabber, kingdom builder, empire builder. What are you up to now, what are you looking at for opportunity? I know you're doing some investing, you have some independent boards, what's your world like now here in Silicon Valley, what's your activities look like, and what's your thoughts on the valley in general and entrepreneurship and your activities? >> First of all, what I'm doing, the good news is I'm sitting here in Silicon Valley and so I'm very busy doing various interactions with bbcs, with startups, consulting, looking at different businesses, there's so much interesting things going on. Every morning you can look at new things that people bring over. Whether they're teleco related or not teleco related. Just some amazing things going on. Something from new wireless protocols to cloudification and so on, and I also sit on a few boards so I'm spending a lot of time doing that, looking at different things. >> [John] What's exciting for you right now? What's getting you jazzed up? >> There's so many different things, like I said, I think... >> [John] What's the coolest thing? >> The coolest things, some of the most cool things are.. >> [John] Confidential? >> Yeah, stealthy, I would say. I'm looking at some wireless stuff that's pretty revolutionary that I think could be new protocols that sort of change the whole dynamic of how wireless works, that's pretty interesting. And then I'm looking at some other things that are just how you apply cloud to different problems in the world. If you look at the cloud paradigm, it's existed for a fair amount of time now, but although we talk about it all day, most of the things in the world, most of the apps, most of the problem sets are not leveraging any in the cloud. They're still at best using >> [John] Recycled IT. >> Recycled IT, or sometimes even Windows 98 for all you know, right? In Africa, Africa went to wireless directly, they never did wired. So there may be a lot of industries that never go from Windows to proper data centers. They just go straight from basic Windows directly to the cloud. There's lots of opportunities that are interesting there. I'm looking at a few CEO options. But it's very exciting, there's so much going on. There's just so many things happening. >> [John] Well let's get into that tomorrow, you're going to come by tomorrow at 4:30, folks watching tomorrow at 4:30 Pacific Time, Saar will be back in the studio, we're going to dig in to the entrepreneurial landscapes, I think one of the things that you highlighted that we were talking about earlier is that sometimes you have technology looking for a problem, and the reality is that most of the game changing opportunities come out of left field that no one sees, these are the revolutionary game changers, the new technology, the hard stuff, not just some app that gets built, it's the real hardcore tech that could be applied to some of these real problems. And I think that's going to be the key. Saar Gallai here inside the studio, breaking down Mobile World Congress with theCube here in Palo Alto covering what's happening in Barcelona. We got some still phone-ins, late night in Barcelona, we're going to make those shortly, be right back with more coverage after this short break. (soft music)