Partha Narasimhan | HPE Discover 2017
>> Announcer: Live from Las Vegas, it's theCUBE! Covering HPE Discover 2017. Brought to you by Hewlett-Packard Enterprise. (electronic music) >> Welcome back everyone. We're here live in Las Vegas. This is SiliconANGLE Media's theCUBE. Coverage of HPE Discover 2017. That's Hewlett-Packard Enterprise. I'm John Furrier, the co-host of theCUBE. My co-host for the next three days, Dave Vellante, Chief analyst at Wikibon and co-founder with me of theCUBE. Our next guest is Partha Narasimhan, who's the CTO of Aruba, now part of HP for multiple years. Welcome to theCUBE, thanks for joining us. >> Thank you. Thanks for having me here. >> So chief technology officer of Aruba, we get to go under the hood a little bit, but it's really important the we kind of have that context because the wireless is beyond just wifi. I mean, wifi is joked on the internet as the lower level of the Maslow's hierarchy of needs. I mean, it's the biggest complaint people have. 'Where's the wifi? Is it secure?' So wifi obviously in wireless, certainly brings that digital life. Now you take wireless in an enterprise environment, unlimited possibilities with Internet of Things, campus, intelligent edge was the key part of this theme. Welcome to theCUBE and thanks for joining us. >> Narasimhan: Thank you. >> So your thoughts on the intelligent edge, what is Aruba's real innovation happening now? Where's the fruit coming off the trees, beyond just the access to wireless, because we all know is like, freedom. >> Right, and so think of it. People tend to think of wireless as just okay, getting rid of the cables. And yes, it does that, but it does a lot more than getting rid of the cables. And if you think about what it allows you to do, wireless enables mobility, right? So as people walk around and stay connected to all the things that they need connectivity to, we allow that to happen. But wireless enabling mobility also means that we have now the ability to observe things in the physical world that people reside in, right? Kinds of things that they interact with, whether it's in the workplace as part of their work, or whether it's as a customer when you enter a retail store, or when you enter a stadium or a venue for an event. How do we now gather our understanding of context from purely from a digital domain? Combine that with what we are learning about context, the other side of context from the physical domain. Blend that together, one, to either create new experiences for enthusers. Or to enhance existing ones that they're already used to, how can we make it work better? Right, and we have these used cases in- >> Is it true enabling technology and these new scenarios are new? I want to get into the used case, but I first want to get the hard news out of the way. You guys have some announcements here at HPE Discover that are notable. One is an extension to the course, which is 8400 line, real focus on the intelligent edge. And the other one is really comprehensive asset tracking. Assets could be anything, it could be people, it could be machines, devices and whatnot. This really kind of highlights the things that are possible. Obviously asset tracking is entering a change, and it's also big data, it's Internet of Things. Talk about this intelligent edge, talk about the switches and the asset tracking. What is the notable things about those two announcements? >> So let's start with the 8400, right it's a big core of new classes, Abdication switch. We don't have many of those built over the years in the industry, right? There have been very small number of instances of switches like this built in here. But the big difference with the 8400 is it's actually the first one that was completely built in an era that is defined increasingly by mobility in IoT, right? Think of mobility. With mobility, topology is never a constant. Right, I can never assume that the network topology is set and it never changes. Or changes can be controlled in some perspectives. So if you start with the given that topology is not a constant, how do we now create a switch that enables network administrators to deploy networks without having to worry about, you know, old methods of configuring and monitoring the switches, right? So, this was built ground up with the notion that it has to support today's use cases, not overlay today's use cases onto, you know a switch that might have worked many years ago. So programmability was important, modernality was important. It is built on a basic premise that you know, the entire state of the network, the switch, is stored in a database that is accessible through modern interfaces. It allows you to take that switch, plug it into existing workflows that are more modern than to try and force feed these modern workflows be constrained by technologies that were invented fifteen, twenty years ago. So that is the one that's important, right? And in addition to all of the connectivity options, the speeds and feeds that we have to go build, it's extremely interesting and there are advances and innovation on that side. The one that excites me the most is the software layer in the switch, right, that allows for the programmability, it allows for visibility, right, I'm able to get access to all of the state of the switch, and any trouble, problems that I see, I can quickly get to the bottom of, and go back and fix it. But more importantly, the biggest innovation that I think we have, is this thing that we call the Network Analytics Engine. This is what gives you the extensibility property for the switch. Where your ability to go extend the capabilities of the switch in the field, right? Not have to for our engineers to tell what features, but people in the field, either customers or our own field engineers, can extend the capabilities of the switch to go do things that we didn't think of, but exist in the field. >> So before we get to asset tracking, I love the way this press release was worded. You kind of hit on it, John. 'Setting it apart from other core aggregation switches in the market, the 8400's most unique innovation is the Aruba OSCX, a modern software foundation that is a departure from today's static, inflexible operating systems.' I'm not sure exactly who you're talking about there, but nonetheless, so when you talk about this network analytics engine, Are you suggesting that makers of static inflexible operating system designs would have to bolt something like that on? And it would just not be as effective? Or would it not be possible? >> Well you have to go deeper into the operating system, and do major surgery for that to happen. And that is very hard to do. So if you already built it for a certain static use case, and you're trying to now introduce support on certain things in there, you will always be limited by what lies underneath, compared to if you look at the operating system Aruba OSCX, it was built from the ground up with the modalality and extensibility in mind including the programmability for configuring the switches. This is not something that you can bolt on later and go fix it, because you have to go very deep into the operating system, and there's a lot of major surgery work that is for a switch of this magnitude and size, the effort is as good as starting from scratch, throwing everything away and starting from scratch and building it new. That is very hard to do. >> And tell us about the asset tracking. Let's not, I didn't want to lose that either. >> Well the asset tracking, before we get that point. Operating systems are generally about connecting things, subsystems. You mentioned the wireless as an unpredictable. Which is true, I agree with that. Now Internet of Things is even more unpredictable, and the rate of devices connecting: a watch on someone's wrist, a wearable, a phone. We saw the new Apple announcements yesterday, we covered that on SiliconANGLE.com, again more and more consumer devices, and yet, industrial equipment. This is important, I mean, it's not just machinery, it's like, it could be the air conditioning, it could be anything. Thoughts on this asset tracking concept? >> Narasimhan: I'll start the IoT's side, and how it applies to the network, and get to the asset tracking. Let me start with an example. One of our university customers, in adding to the story, right, they said one day the facilities people called and said hey, we just went out and swapped out about 400 door locks in a residence hall with wifi enabled door locks, and then they turn on the 400 of them and they don't work, and the network people are saying sure it doesn't work, we're not surprised, because the network's not built to have arbitrary things come in and start working. But if you look at it from the people that are deploying this equipment, they just think hey, it's wifi, it should just connect and work, right? Versus people running the network have certain ways in which they have configured the network either for quality of service, or security, that, you have to understand. But the network admins are also in a state where they just say that battle is probably lost. People are going to add things to my environment and my job is to go make sure that I gracefully accept them in and fit the into a profile. In the security profile that I already have. So this is why the programmability is key, right? As door locks come in, chances are those door locks don't need to communicate with everything on the network. Maybe only to a very specific server that pushes policy on which keys should open a particular door, and which one's shouldn't, but if you look at network's specifically designed with a certain notion of trust, saying okay, I have a security perimeter. Whether it's a physical perimeter that allows me entry into the building, or a digital perimeter that, you know, I have a dmz in a firewall, but once you're on the inside, you're in a safe zone. That safe zone is being challenged by IoT, right? Because these devices are not capable to begin with. Think of an electronic door lock, and compare it with many other servers, and the compute capabilities and the servers on the showfloor, versus what is available in the door lock. By itself, it's starting out at a very huge disadvantage on that front. But more importantly, the compute capabilities of a door lock are frozen in time. And they tend to stay on the door for ten years, fifteen years, twenty years. Versus the compute capabilities that are available to an attacker keeps pace with technology. So over time, the ability for the door lock to protect itself diminishes, right? And this is not just true of door locks, it's true of any of these IoT type of things that are getting added to the systems. So our goal is as we gracefully accept them and get them connected to the network and provide the network guys with the tools to kind of segment them into appropriate boundaries, we also have a need to go monitor their activity all the time. Because they become juicy targets for hackers. To get into and from there, propagate out to the rest of the network. So how do we know they're not getting attacked? Their vulnerability is not getting exploited, and the only way to do this is to continuously monitor them. But more importantly, to use a machine learning approach. A behavioral approach, rather than a signature based detection. So if I baseline behaviors of a certain IoT device, whether it's a door lock or a security camera, or any number of those, and whenever I see deviations from that baseline behavior, I want to alert somebody. Because, while it is an anomaly, and not all anomalies are malicious, but it could be potentially. And you need somebody to look at it, so how do we quickly detect that, okay, there's a deviation and is it malicious, and how do we react to it, and how do we shut it down? >> Now talk about the asset tracking announcement, because that's relevant to this show. What's the status, the news? What's the notable thing about the Aruba asset tracking? >> So the asset tracking is interesting on two counts: one is, when you want to track an asset, you need to attach a tag to it, that kind of puts out some notion of identity that I can attach to that asset. But there's also the infrastructure side of it, you need an infrastructure that picks up on these signals, and can locate the asset, and can figure out where all it's been, and who's kind of used it, and how it's getting used. So while we look at asset tags as one improvement, the infrastructure is also another improvement. Aruba we started down the bluetooth low energy as one of the elements for our location based services strategy, look at Meridian team as a central gravity for all location based services. We started out with beacons, and using the beacons as an indication of location, and then providing contextual information and semantic information about that location. But over time we've also recognized that in addition to beacons, we've integrated bluetooth into our access point. So any wireless LAN infrastructure that's based off of the Aruba solution that is getting deployed today is already ready for one half of the asset tracking solution, so the other half is the tags that we're introducing that can be attached to high value assets, and then together these two combine with a back end solution that not only provides you with where the assets are and where they've been, but also uses APIs to integrate with other asset tracking larger solutions that you might have. >> Furrier: So you're feeding data into a bigger picture on the edge? Alright, so let's talk about the fun stuff now. So we can get under the hood there. Getting a little technical, which is great. There's some real sexy deployments out that you guys have been involved in. One is, we've talked about, the folks watching theCUBE know, Levi Stadium in San Francisco, for the San Francisco 49ers is highly instrumented venue next is the provider, great service provider. You guys are the provider there. But it's interesting, this other use case. So talk about how the use cases are driving the value, and how people who are discovering these new value creation opportunities, how do they get there and give us some examples. >> So Levi Stadium, okay. We tend to talk about it a lot, but it also is because we learnt a lot of things as part of being, you know, we were fortunate to be part of that exercise. We learnt a lot of things, so one key learning was the people that built the stadium were very clear from the beginning, their competition was not another sporting venue, right? It's not the Oakland Raiders fans that they wanted to attract over into the stadium. They were very clear their competition was a TV in your living room. How do I convince a fan to not watch the game on the TV in their living room but instead to come into the stadium and watch it? So they said here's a list of experiences that I want to create for the fan that you can only have in the stadium, and that is compelling enough to get a fan into the stadium to watch the game. So once you define those, then it becomes a lot easier. The rest of it underlying here is all technology, and we know how to piece it together. So this was one of the first instances where we saw the whole infrastructure build out being driven top down, right? You define your experiences and then you work your way down on what kind of pieces of technology we need to go in there. Otherwise the networking industry was largely driven by hey, here's the connectivity infrastructure, and we'll figure what hooks are available, and then we'll figure out what you can do with it. >> Furrier: You constrain, naturally, with that. >> Correct, but if you start from the top, lot's of things become easier, right, what we need can be customized to go address the problem, specific problems if we want to go in there. And this is not something new. As you look at, yes it was true for the stadium, but retail stores have a similar problem, right, how do I convince a user to not buy a particular item online, and instead come into the store to do it? And Apple's done a pretty good job of it, the whole store is not a place where you go pick items off of the shelf, put it in your cart, and check it out. It is a place where you go and interact with the item. >> Experience! >> Experience, again. >> What other verticals are hot right now, obviously retail makes sense, sports makes sense, what other verticals- >> Narasimhan: Hospitality is another one where you want to create an experience that customers associate with a brand. If I can somehow make the whole process of you checking into a hotel room, and being in the hotel room and then checking out, if that whole experience can be made significantly better, then I'm going to associate it with that brand. And the next time I'm looking for a hotel- >> So it's an opportunity for brands to extend their brand value, digitally. >> Narasimhan: Correct. It is by creating a better experience in the physical world, you tie it to your brand, and customers associate it with it. And then you go into the workplace. Again, employees and how do I help improve their productivity with what kind of experiences? More and more workplaces are shifting to more open environments, all wireless workplace, because, you know, even if I had wires, there's nothing that I can plug into. I have three devices with me in my bag. Nothing with an Ethernet port in it. And we recently moved into a new facility in Santa Clarita. in January of this year, and the one thing you'll notice is that there are no wires running to where people sit. So when you go into an all wireless workplace, you convince users that it's viable, right, and the way to convince users it's viable is that the network is always there when they need it, when they need to do something on it. But the facilities people are also excited by the possibility, because it gives them a lot of flexibility in how they want to reconfigure space. >> Furrier: It's the future of work and the future of play kind of converging with analog and digital kind of coming together with experiences at the center, and wireless is the fabric for it all. >> Correct, and wireless is what enables mobility but more importantly location is the other fabric that interconnects the two worlds. If I can sense location on the digital side, location is an attribute of the physical world. That's what bridges the two together. >> Context, interest, location. These are all new variables, a whole new way of doing things. In fact, the modern mobility we suggest on SiliconANGLE, Wikibon and theCUBE is this is going to be the future of work, the future of play, all around modern mobility. >> Narasimhan: Correct. >> Partha, thanks so much for sharing your insights. The CTO here of Aruba Wireless, doing some amazing things, enabling new possibilities here in a new simplified, sounds complex to me, but it sounds like you simplifying, as the new message from HP. Thanks for sharing your insights. Of course, we're bringing that modern day to you here on theCUBE live in Las Vegas, our exclusive coverage for three days. This is theCUBE, I'm John Furrier, Dave Vellante, we'll be right back with more after this short break. (electronic music)
SUMMARY :
Brought to you by Hewlett-Packard Enterprise. I'm John Furrier, the co-host of theCUBE. Thanks for having me here. but it's really important the we kind of have that context beyond just the access to wireless, but it does a lot more than getting rid of the cables. What is the notable things about those two announcements? So that is the one that's important, right? I love the way this press release was worded. and do major surgery for that to happen. And tell us about the asset tracking. and the rate of devices connecting: and provide the network guys with the tools Now talk about the asset tracking announcement, and can locate the asset, So talk about how the use cases are driving the value, the people that built the stadium were and instead come into the store to do it? and being in the hotel room and then checking out, to extend their brand value, digitally. in the physical world, you tie it to your brand, and the future of play kind of converging that interconnects the two worlds. In fact, the modern mobility we suggest as the new message from HP.
SENTIMENT ANALYSIS :
ENTITIES
Entity | Category | Confidence |
---|---|---|
Dave Vellante | PERSON | 0.99+ |
Narasimhan | PERSON | 0.99+ |
ten years | QUANTITY | 0.99+ |
fifteen years | QUANTITY | 0.99+ |
Partha Narasimhan | PERSON | 0.99+ |
John Furrier | PERSON | 0.99+ |
twenty years | QUANTITY | 0.99+ |
Las Vegas | LOCATION | 0.99+ |
Hewlett-Packard Enterprise | ORGANIZATION | 0.99+ |
Santa Clarita | LOCATION | 0.99+ |
Oakland Raiders | ORGANIZATION | 0.99+ |
three days | QUANTITY | 0.99+ |
Apple | ORGANIZATION | 0.99+ |
HP | ORGANIZATION | 0.99+ |
three devices | QUANTITY | 0.99+ |
yesterday | DATE | 0.99+ |
Partha | PERSON | 0.99+ |
Aruba | ORGANIZATION | 0.99+ |
John | PERSON | 0.99+ |
two announcements | QUANTITY | 0.99+ |
Aruba Wireless | ORGANIZATION | 0.99+ |
two counts | QUANTITY | 0.99+ |
Wikibon | ORGANIZATION | 0.99+ |
today | DATE | 0.99+ |
first one | QUANTITY | 0.98+ |
8400 | COMMERCIAL_ITEM | 0.98+ |
San Francisco | LOCATION | 0.98+ |
one | QUANTITY | 0.98+ |
two | QUANTITY | 0.97+ |
first | QUANTITY | 0.97+ |
One | QUANTITY | 0.97+ |
Aruba | LOCATION | 0.97+ |
about 400 door locks | QUANTITY | 0.96+ |
Levi Stadium | LOCATION | 0.96+ |
San Francisco 49ers | ORGANIZATION | 0.96+ |
two worlds | QUANTITY | 0.96+ |
SiliconANGLE Media | ORGANIZATION | 0.95+ |
HPE Discover | ORGANIZATION | 0.95+ |
first instances | QUANTITY | 0.94+ |
January of this year | DATE | 0.94+ |
half | QUANTITY | 0.94+ |
Meridian | ORGANIZATION | 0.93+ |
theCUBE | ORGANIZATION | 0.92+ |
one key | QUANTITY | 0.92+ |
one improvement | QUANTITY | 0.9+ |
one half | QUANTITY | 0.88+ |
SiliconANGLE | ORGANIZATION | 0.88+ |
fifteen, twenty years ago | DATE | 0.87+ |
years ago | DATE | 0.86+ |
400 of | QUANTITY | 0.86+ |
8400 line | QUANTITY | 0.85+ |
next three days | DATE | 0.85+ |
OSCX | COMMERCIAL_ITEM | 0.84+ |
HPE Discover 2017 | EVENT | 0.65+ |
2017 | DATE | 0.65+ |
HPE Discover | EVENT | 0.54+ |
SiliconANGLE.com | ORGANIZATION | 0.52+ |
one day | QUANTITY | 0.5+ |