 from the Computer History Museum in Mountain View, California. It's theCUBE, covering DevNet Create 2018. Brought to you by Cisco. Okay, welcome back, everyone. We're here live at theCUBE, here in Mountain View, in the heart of California, Silicon Valley. I'm John Furrier, my co-host, Lauren Cooney. Our next guest is Riaz Rihan, who's the Global VP and General Manager of Cisco IoT, CUBE alumni back last on at Cisco Live in Barcelona. We got Cisco Live coming up, but we're here at DevNet Create developer ecosystem for Cisco and external cloud-native developers. Great to see you. Thank you, John, pleasure to be back. What's a Cisco guy like you doing at a hoodie show like this? IoT? You know, IoT is so topical, and there's so much interest around it, so just happy to be here with the developers, and just get to meet a few people out here and just be part of this whole event. So IoT, last time, all joking aside, is really hot because you have now the cloud as a foundation, on-premise data, hybrid cloud going on, but the edge of the network is certainly very relevant. So you've got a lot of new things happening. So the question for you is, what industries are early adopting? What industries do you see that are adopting IoT in the programmable way? I'm not so much a sensor network, so they're out there. But as they bring them into the IoT, into the technology IT world, which industries are the most adoptive for you guys? So I talk about a handful of industries that are really leading the charge, right? Number one, I'd say, is manufacturing. We see a lot of activity out there, primarily because for the first time, manufacturers have an opportunity to really converge their data and put it on an IP network, which is exciting. The other big one is energy, both all in gas as well as utility. And then we are seeing huge amount of interest in transportation, both in actually the roadways, as well as the fleet that run on the roadways. In addition to those, I'd say retail and general public sector in cities are big adopters of IoT. So on the IOG side with, say, let's take transportation. So as we know, we know that Uber happened with Uber. They had a death. Now the sensor that could sit argued, latency matters, right? So you've got to have a network to support it. Is smart cities truly happening in a way that's, in your opinion, moving the ball down the field? Where is the smart cities with IOG respect to transportation? I mean, is it still early? And they moving the ball down the field? Your thoughts? You know, I'm asked this question pretty often. And I can tell you that moving a city towards the smart city is actually a massive endeavor. What we find is the cities that are doing this successfully, they kind of start small with a few use cases. Let's say parking, maybe lighting. And then they kind of expand out a number of use cases, but also a geographic spread of where they'll deploy. And specifically, you know, when they work with us, one of the big advances we are making is something called the fog appliance. So we're building this easy to install appliance that can be used in the sections and at various points to enable cities to go smart. So Lauren, you and I were talking the other day about this is that, you know, when the cloud, you bring cloud together with the developers, it's interesting because they have to actually figure out that software can be powering the edge. And, you know, Kubernetes is in one example. And when you start looking at what Kubernetes is doing to the network layer, you say, okay, I got to write software. But most of the edge applications, oil and gas, they're like facilities guys. These guys are hardware people. They're deploying cameras. Well, there has to be, there's software that actually runs on that as well to enable things for people, places and things at the edge. And I think you all have to look at the edge when you are talking about IoT especially. Software, that's the key. What do you say to all those guys that have to relearn software? Come to DevNet Create, I mean, this is a real issue. Yeah. You know, if you look at software in general, right? Software is playing a bigger and bigger role in these applications. Not to diminish the role of hardware or networking or any of the other elements, but software is certainly playing a bigger role. Like, let me give you an example. Let's talk about the fog appliance. You know, one of the things we've been working on quite diligently is building out a single software framework that can sit on a number of different hardware devices depending on what the use case is. And the use case is defined by the customer, it's defined by the industry, it's also defined by the price point. So, what we are seeing more and more job is having a single software framework but being able to deploy across different platforms if you will, and therefore building different appliances to solve different problems. Yeah, one of the things that I think is huge and I want to get your thoughts on this, I think we should do a deep dive on it. And that is that video is becoming much more of a bigger app. We use video a lot, I see with theCUBE, thank you for watching. But there's a lot of data in the video assets. It's not just do the video to communicate a message, there's a digital artifact that's beyond what it's for. There's a, it's now digitized. So, that's now data, your thoughts. You know, when we were talking before, you mentioned video is a data asset, completely agree. What we're finding is, video is now transitioning from just being something that we thought about for safety and security to becoming more of an intelligent asset. Video is also now getting integrated more with the business process. So, let me give you an example. We're working with a manufacturer of Nylon. And this is a process industry that works 24-7. And they're using video to actually monitor the output as it comes out of the machines. Because when the temperature rises above a certain limit, and this is obviously a manufacturing defect, it tends to blob up the nylon which then reduces the value of the nylon from something that's high grade and high margin to low grade and low margin. And video's helping solve the problem. So, the video creates an alert that's part now of the manufacturing process and manufacturing control, allowing management to intervene quickly to kind of at least cut their losses. So, that's an example of how video is now becoming very much a part of the business process, not just safety and security, but well beyond that. This is well beyond surveillance. Yeah. This is more than just normal use cases. So, new value activities are going on with video. Absolutely. The other big one is traffic. And I'm talking about road traffic. Whether you look at tunnels or you look at parkways and so on. We're now seeing video being used to monitor the flow patterns of cars on highways and on parkways. And then not just using that to predict traffic jams, but in some cases predict accidents. Because once you take these data labels and data assets through video and compute them, it's a stream of information that can be analyzed mathematically using an algorithm. And then fortunately, we're able to now use that to prevent accidents potentially, right? So, that's the kind of thing we see video in. And this is just the cost. You'll see a lot more of use cases where video and IoT get very integrated. And again, very happy that Cisco is leaving the charge for that. So, Lauren, I want to ask you a question because I know you and I have been talking about this. And that is that the developer role around this is not obvious. It's obvious to say, oh, we've got to write software. But now you've got to create ecosystems. So let's just say this is one integrated video. They have a by-build decision to make. Do they build it from scratch or they integrate it in? So if you take Riaz's next level of conversation is video is a service. It's a microservice. If it's a data asset, if you believe that, it's a microservice. So it's not trivial now. You've got to figure out how to codify it. What do you guys think about that? You guys are in the experts in the software area. What's your thoughts? So, you know, I can jump in. I think this is an important trend, especially if you look at three industries that I personally work with, manufacturing, you look at energy, and then you look at retail. These are three industries that I think are leading the charge on how they're using video in this context. And how the videos actually provide it, I think is less important. What's more important, as you said, is the microservice that has video as a component and then consuming that and integrating it with the whole value stream within that industry. The other important element, I believe, is the use of video in conjunction with other types of sensors. So let me give you an example. We're working with a large telco, and they have these cell towers placed all across the country. And they actually have video to monitor the cell towers. And that's great. But the problem is, it gives them a lot of false positives. So they solve the problem by using human form recognition. Still, I mean... What's a false positive? A false positive is, you know, a leaf branch blowing. Oh, okay. And that gives you a positive reading, right? Now, they've kind of used, they've used some technology and they've reduced that down, but they still have too many false positives. So they decided to combine the video feed with some of the sensors that they have. For example, when someone tries to pull a copper plate from the cell phone tower, there's a sensor that tracks that. So now combining the video input and the sensor input, they get much fewer false positives and are able to take action much more expeditiously. And co-occurring incidents are a huge, huge opportunity for the IP. So the thing I want to ask you, because I think this is much more business oriented, so I want to get your thoughts on it. Okay, video is a data asset. People say, I believe that. Now I want to operationalize that in my company. We're talking about a new process improvement. That's hard to do because they've never done it before. How are you guys engaging customers? And what are some best practices to get them to operationalize a new, not just new technology or service, but actually integrate into a pre-existing or changing value chain? One of the things we do, John, is we'll engage with customers to do what we call a value management analysis. So we actually sit down with them, work out what their existing process looks like, what an improved process might look like, and importantly what kind of cost they can take out of the process, out of the system, or what kind of new value they can drive for their customers. So it's either an increase in revenue, it's a decrease in cost, or an improvement in process efficiencies. Once we've done that, it really allows us to then pair up that new process with our technology and then actually track how much of the value they've received. We've found this approach kind of grounds everything in a very strong ROI. So instead of guessing as to what the output will be and does it actually move the needle on a value basis, you're actually able to document that up front and then actually track the results against what we thought would be. The other advantage of this process is it allows us to improve incrementally. So the first version of a video-enabled business process might give us a certain amount of value, but as we improve on that, we could see incremental values and other processes being added on. Very similar to starting small and then adding on creatively kind of a design way. But you've got to be open-minded. Let me throw a wrench into the equation here, which is, okay, new data source. You mentioned the co-occurring identity on the self-tower example. It could be like multiple data inputs that are new. How does the customer figure that out? Different customers are different. And again, as I mentioned. Very generally speaking. You've got to be prepared for the unknown. Yeah, some industries I think are more open to this because they've seen, they've felt this problem before. Going back to manufacturing as an example. Monitoring, visually monitoring, the output of a manufacturing process is a very labor-intensive proposition. Manufacturers have struggled doing that for a long time. Now, having used video and getting just a very high level of efficiency and combining different types of inputs is something they're very open to. So we see them very open to it. Other industries, I think, are coming along, but it all goes back to how important a problem are you solving and what's the payoff of solving the problem? The bigger the problem, the more the willingness. Great conversation. We had the DevOps guy on early Damian who's with the run deck and he's saying, you know, breaking down silos and tickets is killing operations because of an old paradigm. You guys are going down a new road and we talked about this at Cisco Live in Barcelona. You got the tailwind for you guys. But you got a clean sheet of paper but you got some pre-existing stuff but it's not like baggage. It's just an opportunity, might be. So I got to ask you, how is the business going? What are you guys doing? What are some recent successes you've had? Share some insight into that. Yeah, and I think, just to add to that, I think, what are the revenue opportunities that you see that, you know, you're providing these services to these customers? They must see new revenue opportunities as well. Wondering what those are. So I'll kind of cover both sides of my business. I'll start with Jasper, Cisco Jasper. Our Cisco Jasper business is doing fabulously well. We added almost 10,000 devices last month along or last quarter and we're on track to keep adding devices at a very fast pace. So we're very excited about that. We've just crossed 75 million. So 75 million devices on Jasper. The last time we spoke, John, the number was 60 million. In addition, we've also seen many more enterprises adopting Jasper. The last time we spoke, it was 14,000. Now it's over 16,400. So that number just keeps growing. We'll see you next month, it'll be 20,000. So that number just keeps growing and that allows us to broaden our reach, get into different use cases and drive incremental value for our customers. On the Kinetic side, as I mentioned, we're seeing a lot of traction in the verticals that I laid out earlier. But specifically what we're finding now is customers are getting serious. Take a minute to explain Kinetic for a second. Just one minute and then get into it. So Kinetic is our data fabric. It's our platform that allows us to extract data from all kinds of IoT devices that are sitting on a corporate or a private network. It allows us to process that data at the edge and then it allows us to transport the data to wherever the customer wants it to be. So it's really our IoT platform, our data fabric at the core. The Kinetic business is doing great. We've had lots of uptake. There's actually a booth out here where they're demoing Kinetic. I see a lot of people coming in and trying to understand it. And we see people deploying Kinetic in more and more unique ways. We're working, for example, with a German manufacturer, a very prestigious German manufacturer that's now launching a pretty large project with Kinetic where they're using Kinetic to monitor the health of not just all their new machines, but also all the brownfield machines that they have installed over the past decade. So we're very excited about that and very excited about the future. Well, great job. Congratulations. Always great to talk with you. I think it's one of the exciting bright spots within Cisco with the IoT. Certainly the DevNet developer program has been a huge success and that's only going to help you guys and also the DevNet create. You want some more software developers working on Kinetic and also Meraki and all these cool tools. So congratulations. Thanks for coming on. More live coverage here. DevNet creates theCUBE in Silicon Valley in Mountain View, California. We'll be right back after this short break.