 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. We're back here at Dell World, Dell EMC World in Austin, Texas, 2016. This is theCUBE, the worldwide leader in live tech coverage. Andy Rhodes is here, he's the vice president and general manager for IoT at Dell EMC. Andy, good to see you, welcome to theCUBE. Great to be here, thanks. So IoT's hot, it's the next big wave, you guys are riding it, give us the update. Yeah, it's definitely at the top of the hype cycle right now, that's for sure. So we get a lot of calls and people ask, what is it? That's the first question, what is IoT and how do I buy some, how can I buy me some IoT, is there a skew for that? And I jest a little bit, but the reality is that it's really a series of use cases. Everything from the consumer wants that people know, like a Fitbit, which is monitoring you as an individual and taking sensor data off you and pumping it to the cloud, all the way through to things like predictive maintenance on heavy mining equipment, through to cold supply chain to energy demand response. So IoT is really just a series of use cases where our customers and the world is collecting data of things rather than just people and applications and then really analyzing that data for some sort of business value, whether that's driving efficiencies, if you're doing building management with smart lighting and smart air conditioning, you could be wanting to save electricity. If you're doing, the smart hotel room of the future, it could be for the customer experience, no keys and great environments. Or where it gets really exciting is when our customers are starting to use it to generate new business models and new revenue streams. And so a lot of them, it's all about this digital transformation. We have one customer who is in a building business and they used to build pumps and valves and thermostats. And now what they're doing is they're using some of our products and services to build a building appliance. They sell as a monthly fee and manages the building as a service. And then the next stage in their evolution is now, if they manage hundreds or thousands of buildings, they can compare those buildings together and they can say, hey, look, I got two strip malls in Houston, they're both retail. I can evaluate the use of energy in that building. I know it's because one of them has an old air conditioning unit. I can now sell that data to air conditioning resellers in the vicinity of those shopping malls. So that's where it gets super exciting is this business transformation effort. I want to test a premise because when you hear things like that type of transformation, you think disruption. And you're talking to a lot of customers. When you go into any retail organization, they've all got an Amazon war room, they're getting disrupted, they're getting Uberized, they're Airbnb, okay. The premise I want to ask you, put forth is, it feels like IoT is transformative, but not necessarily disruptive. It's the existing infrastructure players that are taking advantage of that. Is that a fair premise? What are you seeing? I've become known for this saying in the last 18 months of this job, which is it depends, right? So it's my new buzzword. It depends. You should become a consultant. It depends. And so I think that it can be disruptive. I mean, look at things like Uber and Airbnb, they've become verbs today. You've been Ubered or you've been Airbnb. Well, I mean, there's many out there. I find that massively disruptive on the business model side. But there's still this sort of transformation that is everybody else in the world. So if you're running a factory and you want to make it more efficient and smarter, then you don't have to throw away everything you've got to get there. Sometimes it's just unlocking these islands of data that already exist, but don't sit on an IP network. And so when you unlock those islands of data and then you can mash that data sets together, then you've got new use cases, you've got new ways to analyze this data and all of those sorts of things. That's more transformative. So I think there's both in the environment. Andy, can you drill down a little bit for us and where does Dell fit into this big ecosystem? The hardware you're selling, is it solutions, software, all the above? Yeah, so I think maybe I'm biased because of the company I work for, but behind every IoT project is an IT project. If you think about it, it's about extracting data and analyzing data. The first problem statement is much around connectedly unconnected. Most of the things in the world already exist. They are production lines, they're elevators, they're air conditioning units, they're lights. They exist out there. No one's going to throw them away just because they got religion about smart environments. And so the first job Dell does is connectedly unconnected. And that's why we set up a new division about 18 months ago around IoT where we're actually building what's known as an IoT gateway. And an IoT gateway basically connects to multiple different protocols. Whether it's old, serial, like RS485 or 232, or brand new stuff like ZigBee and Z-Wave and other technologies, we can ingest all of that data from multiple of those protocols. Second part is, well, how do you analyze and where do you analyze the data? So if you think about this notion that all data will go from sensor to cloud, to public cloud, that's a very consumer-esque mindset. And we know from hundreds of deployments now that that's never the case. And so, and not all the data has value all the time. And so you've got massive data sets. If you had to ship them back into the cloud on sort of cleanse, you'd have just this massive data. It's not going to happen. Yeah, no way. Our CTO, David Floyer, has actually written extensively about this because data has gravity and just the sure physics of what you're talking about. There has to be a lot of edge computing in what happens with architecture. Yeah, and this is predicated on the fact that 82.7% of statistics are made up on the spot, but I'd say that 90% of that data is useless in its raw form. So I see the gateway as one piece of the architecture and it's almost like the spam filter to the cloud in an IOC setting, right? Think about it, it's cleansing data on the way, extracting the valuable data. The second part is you do have a lot of data back in the branch or in the data center and you do need to do big data analytics on that. And then Dell, especially now with Dell EMC, we have massive of capability to store and analyze big, big data sets. So we're the underlying infrastructure provider on all of this, and then we have great partnerships with platform vendors, whether that's Microsoft and Azure or SAP and their platforms or PTC with ThinkWorks. I've said it many, many times, publicly and privately, anyone that will tell you in this IoT world that they can do everything for you, they're a liar. And so partnerships and a really strong ecosystem base are absolutely critical. So where we don't have the capabilities, we partner, and then we bring all of that together in what I call use case blueprints. So here's how you do predictive maintenance on heavy equipment. Here's how you do cold supply chain with perishable goods. Here's how you do automated energy demand response in factory environment. So I think that's what the world needs is more of these how-to to a level of specificity but not all the way down because we don't know what sensors they're using, what data they're collecting and every one of those use cases. And then you come in with a solution somewhere in that value chain for that use case. And give us some examples. What do some of those solutions look like? Well, predictive maintenance is a great example. So you look at people that are operating heavy equipment, then what they tend to do is they have to maintain that. They're doing it right now on a time stamp basis. Imagine mining equipment in Calguli in Australia, so a big mining area. While you're in the middle of the desert, you've got heavy equipment, maybe shake and bake, separating rocks from other rocks. That stuff breaks all the time. A lot of those manufacturing equipment providers today, they will go service that every three months, whether it needs fixing or not. You then wire all of that up with sensors and you can start to extract the data from that and then do sort of predictive maintenance algorithms and you know when it's going to break and then you send a man or a lady in a van or a helicopter in many of these cases at the right time, not just on time intervals, and you can save a heck of a lot of money on your own bottom line because you're not shipping people around the world all the time and you can keep your customers up on whatever their environments are. So that's a great use case around predictive maintenance. You were talking earlier off camera about the cultural issues. Talk about OT versus IT, the operations folks versus the IT folks. Do they talk? Should they talk? How do they converse? What's that cultural divide and how is it coming together? Yeah, this is actually one of my, the cultural side of this is my biggest learning and one of my favorite topics because very quickly you realize the tech is there. So the tech's not that hard. Well, don't do it yourself, we can do it for you but it's not that hard. But it's available today. It's available and it's commoditizing and the price is there. And it's not a mystery anymore, right? It's not a mystery. Now, connecting some of these things with the non-IT protocols is tough but the industry's solving that problem. No, the OT and IT, OT by the way is operations technology. OT people don't call themselves. You won't find that on a business card. They'll be called manufacturing VP or head of logistics or fleet management. Facilities. Facilities only, yeah. And their world is much more of a business world. I mean, they are the business. Their metrics are the yield of my factory, the yield of my agro crop, their, the energy efficiency in my building, they're the uptime of my manufacturing process. They're very different metrics where if you look at IT, their metrics are more around security and scale and cost of deployment. And so these two people have to come together because my view is once data hits the IP network, it is the CIO's responsibility to secure that data, to make sure that it gets used or thrown away in the right way. And the CIO, she or he, is absolutely responsible for it. But it's useless on its own, right? Data stored is a cost center, data analyzed as a profit center. And that's where the OT people have to come back in and actually make use of this data. They're the ones that say, hey, I'm going to use this data in a certain way, but IT already have big data, you know, analytics systems. They already have sensors of excellence. So what we've seen in the most unsuccessful deployments is where IT kind of sit around saying, I need to have an IOT project. I don't know what that is, but I need one. Or OT say, I'm going to go do it on my own. They have to come together, work together, understand what the use case is first, understand what the value proposition is and what the business value is, whether it's new revenue stream or efficiencies, and then they have to work it together. So that's number one in the cultural interlog. Then if you think about, you know, a lot of customers are using this to change what they sell and how they sell. So IT and OT can come together and create a new product. But if your sales people are trained on how to sell maybe an OPEX model versus a CAPEX model and your marketing team haven't caught up or your CFO says, I don't know how to count that revenue or I'm going to see a revenue dip for two quarters, you know, how do I explain that to the street? So I'm now seeing that successful IOT projects start with the use case, have OT and IT come together and then bring in the rest of the business functions. Andy, one of the biggest topics we've been talking about recently has been security. So, you know, what do you say to people that say, well, IOT is going to greatly increase the surface area. We've got all these news protocols. It's highly insecure. Oh my God, just stop. Yeah, I mean, look, the first thing I'd say is that's absolutely, you know, getting paranoid about security is the best thing you can do. So getting paranoid to the point of not doing anything is also not the right thing to do. And I think that the industry is coming together to help solve a lot of those problems. Yes, the surface of attack has grown, but there's ways that you can set up, you know, your network protocols. Using a gateway is a great example of starting to bring all the data into one device that you can secure. You can lock down the ports. You can manage it like a server. You can put, you know, network security. You can put encryption on the data. You can do all of these things and it's in the IT's domain and they know how to do that. Now, you also have to partner with some of the OT security vendors. GE, as an example, has got some great assets around OT security with WellTech and that whole portfolio. So I think like anything on security, it's not one thing. You know, it's not one security solution. You have to do it at the things level. You have to do it between the connection of the things and the gateway. You have to secure the gateway. You have to secure the data going from the gateway into the cloud. You have to like just deploy good network security protocols as well. You know, one of the ways that you help security is not do dumb stuff to begin with. And so, you know, you've always got to be paranoid about it but if you're paranoid and paralyzed, then you're going to fall behind the world very, very quickly. So again, I think it's coming to a company like Dell who understand all of those aspects and bring in the right partners or our own assets to help customers. But you know what's interesting there, Stu, is the model, we always talk about it's not about the perimeter. It's not about building the mode. It's about, you're going to get infiltrated. It's about using analytics to find out who, and what to do about it, how you respond. Now, while that sounds good, a lot of organizations still put most of their investment in the perimeter. With IoT, there is no perimeter, right? I mean, there's a series of perimeters, right? And so, and I feel like Stuxnet changed. I mean, that's a good example of an IoT worm, if you will. That sort of changed the mindset. So, are you finding that customers in your world are more tuned to this reality? And are able to better respond, or are they not as security savvy as say, traditionally in paranoid as the traditional IT folks? I'd say that most customers, whether they do IoT or they do traditional IT, given the high profile of attacks, given the catastrophic business losses, the consequence, I'd say that 95% of the customers I speak to, it's one of the top things in the dialogue, is how do we secure the stuff, and how do we manage the data, how do we secure the data? I haven't seen a lot of them paralyzed by it, to be honest. I think they realize- It's not, it's safe. No, I have not. That's good. I think they realize that the advantages of going off and putting these use cases to work in their business, and the fact that the world is helping with the security issue, just look around this show and you can see a lot of security providers. I think that they aren't paralyzed by it, but they remain very, very paranoid. And it's the mindset that they need to have and continue to have, because the last thing that they want is an attack that will stifle the first or second IoT use case project they put in place, and then it all sort of breaks down. Right. Andy, we talked about how all the data doesn't necessarily go to the cloud, we have the edge component, but what about the ties to the cloud? Dell, for the most part, does public cloud through partnerships, so how does the public cloud fit into the overall IoT story for Dell? Yeah, I think Michael articulated this, and I hate to paraphrase him incorrectly, but I think he said that the cloud computing is a way that you do computing, not a destination, right? So when I say the cloud, I think it depends on what the customer's requirements are. Some of them just, they absolutely cannot have their data go off-premise for security reasons or latency reasons. Some are fine with that. Some it's for national boundaries, and so we have a very open approach to that architecture. If you want to go and do computing in the public cloud, then we have partnerships to do that. If you want to go and do hybrid and put some of it on-premise in what's now called the fog, we have a new term, it's the fog, so you have the edge, which is the gateway, the fog, which is sort of the branch or the middle, and then the cloud, we can do it that way. I come back to this terrible answer of it depends, and it's use case specific, and it's company specific, and so if you're only a cloud vendor, guess what, everything looks like a cloud. If you only build edge gateways, guess what, everything looks like an edge gateway. If you only build the stuff in the middle, that's what you sell. The beauty of Dell EMC is that we have that whole portfolio, and we can actually sit down and advise customers based on their needs of what is the best solution for them, and it sounds like a sales pitch, but I firmly believe that, and if you go speak to our customers, that independence of architecture is hugely valued. It's all open. Now, we're always going to recommend that you buy it all from us and that our stuff works better together, but if you want to uncouple it, then you can uncouple it because it's based on open standards. And it's a great topic. We got to leave it there. Thanks very much for coming to theCUBE. Really appreciate it. Great hanging out, great chatting. All right. Keep right there, buddy. We'll be back with our next guest. This is theCUBE, we're live from Austin. Right back.