 Welcome back, everyone, CUBE coverage here at Cisco Live Day Two. I'm John Furrier, Dave Vellante, host of theCUBE, industry analyst, breaking down all the action. We've been here for three days, counting the preview day, and what a great change to see a platform focused, unified across all of Cisco, building a bridge to the future, and our next guest was on stage for both keynotes, Jonathan Davidson, executive vice president, general manager of Cisco Networking, core part of the business, good friend of theCUBE. John, thanks for coming on theCUBE. Big week. Thanks for having me here. It's a big week. One of the biggest and really excited to talk to you about it. The bar is really high now. The story's so good. Yeah, the sizzle, the steak better be good. Let's see, this is a great event. I think you guys really nailed it. The 13 years of covering Cisco, watching them, you know, big company, really the most valuable company in the world, and then the way the world changed and grew with the internet, a lot's gone on and a lot's changing. And now it seems like we're at an inflection point for Cisco as a company. Not to move up the stack, everyone talks about it, to have a platform and your theme was unification, unified experiences, building a bridge, but the networking is powering everything. The connectivity is the core data and networking of the hottest areas that we're talking about in theCUBE. At every event we go to, and that powers machine learning, cloud computing, the way we work. This is your wheelhouse. This is like you're in charge. I think Cisco has always been the epic company for networking and now security and networking go together. You can't think of one without the other. And obviously we want to make sure we've got that simplified experience, but you talk about the sizzle, where's the steak? You know, we were really careful to make sure that we actually had products that were out of the market already so that we could show proof points towards this unification. And we started hinting at it last year. We started talking about cloud monitoring for Catalyst, but we knew we wanted to talk about the broader platform. We knew that we wanted to talk about this broader Cisco networking cloud, but we held it, which was hard for a whole year, so that we could have more proof points showing people that, hey, we are going to simplify the experience for you whether you're on-prem or whether you're in the cloud. And a lot of it came out of an organizational focus too. It's tight right now. You got networking. You get the groups highly focused. And then everything else goes into this new group to call this new non-new group. They renamed it OutShift, which is the emerging technology. So I call that the test kitchen, but they're working on new businesses. So you have the core businesses all platforming up, networking security, and then they can differentiate on their own, and then the other group. How does customers reacting to this kind of change? Because it's multi-year. Absolutely. And what's their response been? The response has been phenomenal. And we have met with a lot of customers starting with our global customer advisory board. And we've gone from there to lots, we talk to lots of customers every single day. And we've been putting this message out there for months now. And the feedback has been, okay, this is great. I want my team to meet with your team because I want us to align so that we can start moving aggressively with you. And that was exactly, we couldn't have hoped for any better feedback. And the reason for that is because if we go through and we drive this unified experience, but we get there, our customers aren't with us, there was no point in the journey. So we really want to make sure that we make this journey seamless, simple, and as easy as possible for them to move along with us. I'd like to explore that a little bit and sort of what's under the covers. Cause you're going from this sort of product mindset to a platform, you guys talked a lot about features to outcomes, okay. And we hear this cross-cloud, multi-cloud, what I call super-cloud, John and I call it super-cloud, but it has meaning, right? It's not just multi-clouds, maybe what multi-clouds should have been. So what's the enabler there? My understanding is you're creating a consistent experience. You just said that on-prem, within any cloud, across cloud, potentially out to the edge at some point in time. What's the enabler there that technically allows you to create that experience? Well, if there was one enabler, I would have waved that magic wand a long time ago, but it is a whole set of things. So for example, getting a consistent experience means that you have to have a consistent way of developing your products. You need a common UI-UX framework. And so GSTOOS team and my team and Liz's team, the engineering teams all came together and defined this. We have an internal name for it. I'm not going to share it with you, because it's not relevant. And we're all, we're all snapping to that framework. And we're product by product going and bringing that framework in so you can have consistency of experience across each and every single one of our platforms. And that was kind of step one. And we've been going at that for a couple of years now. That's a software framework, right? That is a software UI-UX framework, exactly. So that's number one. But then it's also how do your customers perceive the platform? So we know SSO. We had five different SSOs. But that's not really what our customers want. You need to have a single sign-on. I always like to joke. It's in the name, single sign-on. Why do we need five? So getting convergence around that is really important. So I log in once, I'm logged in everywhere. And guess what? Even if I'm switching applications, it doesn't feel like I'm switching applications. Then it comes down to branding. Like we've made our branding very confusing over time. And we've been simplifying it. But for example, like Catalyst SD Win, formerly known as Viptella. It's going to be Catalyst SD Win from now on. So there's Meraki SD Win. There's Catalyst SD Win. But we want to unify those experiences well. So think of it, if you want really fancy and tint-driven SD Win, we have Catalyst SD Win. If you want a simplified SD Win, you can do that. But if you want to do security with it, it's the exact same UI to add security to the Catalyst SD Win as it is to add it to the Meraki SD Win today, right now. And so it's a whole combination of all these things coming together. And it's just great collaboration between the teams. I was commenting yesterday on our keynote wrap up and review that after sitting through the keynote, you and G2 and Liz and Chuck kind of kicking it off, I kind of felt like that Steve Jobs video when he first took over Apple. We have too many products. We're going to simplify. We're going to go back to Chayate Day. He's very famous on YouTube. But that really was of a structural change where he went into Apple and said, we're just going to simplify and focus on our core. That's what it feels like here. And even the announcements, it's got a little AWS vibe to a lot of flow of news, but product excellence, but not a feature speeds. Even though you guys liked that, as Chuck said, it was very much Apple-esque vibe. And I mean that as a complimentary way in the sense that it seems like simplification. This is by design. Do you take us through how you think about that? What's going on internally? I'm sure pulling people through that change is a management task. What's your, take us through some of that experience. Well, if you go back to even, let's say, 15 years ago, we would be happy if we built a feature, we won the deal, and they didn't deploy it. We're like, oh, it's great. They don't deploy it. They're not going to call us for support. Fantastic. We literally, no one thinks that way anymore. No one talks that way anymore. That would be considered a waste of time. And what you have all equally amongst us is we all have the same amount of hours every day. And so how do we go and make sure that we're actually building to our customer outcomes and start from that way in? And this is why we've started with what's working already. We've been using it for years on the Maraki side. It's this double diamond development approach, right? Which is you go through and you make sure you start with the outcome the customer needs, and then you go through design and development process. And that can take three, six months for stuff you already know. We know how to do routing. But if you're going to build something unique or some new outcome that the customer wants, it still might take three to six months to figure out what that MVP is going to look like, what the experience is going to be. And then you start writing code. And then you actually go and trial it a few times. You build a few prototypes. You go back, see, is this what you're trying to solve for? And then you actually go and build the product out. And so it's a very iterative. But what happens is when you ship the product, it can be adopted that same day versus shipping the product and then having to wait 12 months or 18 months until you get to real MVP. So we don't celebrate FCS. We celebrate adoption. Talk about the impact of networking. Honestly, it does affect the experience of people's, whatever application they're using or workload that they're running, or chips that they've bought in their devices. As network got more complex and diverse with the internet, you mentioned that on your keynote that the internet, people now use internet for backing through a routing. I talked about it yesterday. What should people know about right now that you guys are doing networking that's different than just five years ago or a few years ago? Yeah, well, there's so much that's changed in the last five years. I mean, we have been looking to simplify the portfolio. We have been looking to get more consistency about how we approach our customers and their demands. And the, well, first of all, networking is exciting again. I'll just put it that way. And the reason why it's exciting is I think is one of the benefits, the silver linings of the pandemic is that everyone realized the criticality of infrastructure and they realized this is something, hey, you have to invest in or they will slow down my movement towards digitalization. And every business is a digital business. Pizza chains are digital businesses now. Now, their product is not internet, but if you don't have a great experience for your customers, it doesn't work. And you heard it directly from this morning talking about to the NFL's head of security, like football is not a Sunday event or a Monday event or a Thursday event. You need to have constant integration and collaboration and bringing your customers in all the time. And the only way to do that is through a compelling digital experience. And that requires you to really go and focus on all of that infrastructure. And by the way, I liked her comment. She was great by great interview by the way on stage. You guys did a good job. What jumped out at me was that comment she said physical and cyber security is completely converged for her. And I think that's a new normal. What's your reaction to that? What would you say to the folks watching that haven't gotten there yet, that that converged cyber and physical security, whether it's your device or your office or whatever? Well, it's not only physical and virtual coming together. I would say that when you look at that from a sustainability perspective as well, so the ability to look holistically about making the office somewhere where you want to go to instead of somewhere where you have to go to and also with what we're having with all of the power costs increasing so dramatically around the world, moving to PoE-based lighting can save you up to 50% of your power costs. And so when you're rethinking about how you can build out your infrastructure in the office, you can create a better experience. You can base it more holistic experience. You can enable where you can actually if you're in a new office, if you're a traveling salesperson, how do I find the open office? And instead of wandering out the old days where you open the door, oh, close door, open the door, close door, right? It can be frustrating for all of these mobile offices and you need to create that seamless experience and that is what she was talking about. How do you embrace this and create the experiences but make sure you also know that increases the attack threat surface. So I think she mentioned 2,500 plus screens in just one of the stadiums and if any one of them get hacked, I mean that would be a very bad day. So networking's getting more exciting and it's just going to get more exciting going forward. You're talking about the double diamond approach and the time it takes to get to MVP. Do you think AI will compress that time to MVP if you started sort of applying it for that specific purpose? Yeah, I think there's a couple ways that AI could help. First of all, if you look at the developer experience, so being able to go and get enhanced tooling that can, you put in the comment of what you want to do and it actually proposes the line of code that you need and it's actually pretty accurate for kind of non-C, non-C++ type of types of things but it's going to get better over time especially as you're seeing this. So the ability to, I've talked to a, I'll just say a top Fortune 50 company and one of their engineering directors and they said that their staff was 70% more productive over the past several months because of those, put in what you want, it proposes it, you hit return and you get to go to the next line of what you're trying to accomplish or it could even give you a whole set of code not just one line. So engineering efficiency can go up and that can increase velocity and hopefully decrease errors and so that's one part of it. And then the products themselves, as we move forward we think G2 was talking a lot about prompt-based interfaces. So visually you want to be able to see what's happening across your infrastructure but if you can ask something, a question and then kind of narrow things down which switch has the most errors on it in my network today? Okay, here you go, here's it. Okay, take me to that switch. What do you think the problem is with that switch? You could actually work through these types of scenarios where today that would be a bit of a daunting task if you've got 10,000, 20,000, 50,000 switches or even 20, right? It could be a daunting task. You could probably get to the point maybe you're even there today where you can prioritize the severity of those issues and triage. So many opportunities. And so I think there, I think UI is here, is this because I think some of us are visual learners, I'm more of a visual learner. And so I think being able to see things is always going to be important but having that as an augmentation I think is going to help a lot. I love how you guys put networking plus security as kind of like the pillars. As we know, security is a data problem too. Not a lot of data conversations here. I mean, I know how it's cleaner with security because it's tighter. I like that. I'm not disagreeing with it at all but we all know data is underlying both those categories. And then each group has differentiation on their products. I love that's a platform feature. What's your view of the data? It's what's going on under the covers. How are you thinking about data? Obviously AI is going to help. How are our customers thinking about how they leverage their data? 100%. So this is a, just like it's a, it's a war for talent, always. And if you have the best talent, you can usually build the best products. So we're always focused on making sure that we've got the best talent so that we can go and build the greatest products. What you just shared is it's also a data-oriented business. Networking is a data-oriented business. But if you look at simple facts, like we block more attacks in a day than Google searches, right? That's, there's a pretty, and this is a number that we've shared for years, right? And there's numbers. You can tune that model up. I mean, that's a new product. You can tune that up for, that's your SOC assistant, is that? This is actually through the OpenDNS product umbrella. And the ability to go and do that. But that's just one piece of it. There's a lot of other things that we're doing and blocking on a regular basis. Look at what thousand is. Thousand eyes is great at what it does because it consumes so much data from how else do you know what's happening in real time across BGP of the whole internet unless you are taking in multiple feeds from all over the world of the internet in real time and able to actually understand what's happening. So you can see that IP addresses or entire ASs are being stolen. So you can actually understand is that real? Is that not real? You can send alerts. But then on top of that, now being able to feed an application level data into that view means that you can have more of a digital experience monitoring. So app D, the full stack observability, that is a data-driven business. You know, and that's just very nuanced. I'm glad you brought that up because I wanted to bring this up. What you said on day one on your keynote ties to exactly that point. It's very nuanced. I want to make sure we get it out on the camera. You said network at AI scale. I was thinking AI at scale. Why is AI in there? Oh, data scale. But what you're getting at there is the way the network is with thousand eyes and even just the breached numbers you tossed around. It's just an example of many other data points. There's so much data at scale. That's AI available. So it's almost AI, that's what you're talking about AI at scale. Is that kind of what you mean network at AI scale means the network is scaled up and now a new set of circumstances is coming to the table in AI opportunity. So I think there's two things there. One is we are helping hyperscalers with the largest types of AI infrastructures to build out those infrastructures. So that's the physical infrastructure at AI. The second part of that is around you need the data sets. And if you go back to Cisco's business model 20 years ago, we built a router switch. We kind of threw it over the wall. Someone would install it. And then we would never hear from that router again. The old prodigal son analogy. They only came back if something broke and they would call tech support and we'd find out, oh router, here's the configuration. Here's how you fix it. And then, nice business. They disappear again. But what we really need to be able to do is we need to be able to get real-time data. And if the customer's willing to share that real-time data with us, then we can do a whole host of amazing things. I can tell you when your optic is going to fail. I can tell you that if those boxes are overheated, that the mean time to failure just actually decrease so that the box's likelihood of failure because you had a data center cooling thing that went out and for some reason the box didn't shut down like it was supposed to. And we could see that, we could know it, and we could let you know that, hey, we should pay this kind of attention to these platforms over time. That's all data-driven. And that's because you know what happened, you know what went wrong and you can predict the probability of it happening again, is that right? But we can predict it like things fail. Like hardware fails. We know what the MTBF is for every single component on every single device. We know which batch of silicon every single component went into. And so if there was a bad batch of 100 components, we know if they're telemetry connected, we know where they are, we know what they've done. And we can say, hey, that bad batch is running in this network and we can replace it or there's a 5% higher probability that that device will fail than a normal device. What do you want to do about it, Mr. or Mrs. Customer? And is that AI or just good statistics? That's, well, I think a lot of people confuse ML, stats, and AI pretty often. So I think everyone just calls all of that AI now, which if you want to have a wholly separate conversation about that, I would be happy to. But I think everyone just calls it AI, which is not technically accurate, but we can argue about that or we can just talk about what it does. So let's call it AI. So by the way, my follow up question is you talked earlier about being a visual learner. And it seems to me that so AI is a visual learner as well, it will be. So the more of, and networking is visual, the more of visual props and you can give to the AI to learn, I would think it's going to make it more powerful. Does that make sense? And is that sort of in the plan or? The way I think about it is this. These models, they love structured data and networks are nothing more than graphs. If you think about it from that perspective, like mapping out graphs and if you're able to map out very large graph data models and then apply the right heuristics to each one of them, you can do amazing things with that. And the neural network side is coming online too. It's an amazing thing that we are going to be able to do. So this all requires data. So having that data through a large number of our customers now, enable us to see what's happening in real time. When they do, we actually can be a lot more proactive with helping them understand how to keep their networks up and running to the highest potential of that infrastructure. Is there a data corollary there in terms of like a graph database or a graph data corollary? I think we're looking at how all these technologies can come together. I think it's the way to think about it. Foundation models are hot. Certainly you guys have a lot of action there. One of the things that we talked about a lot with you in the past few cube interviews you've been on is chips, silicon. And silicon and networking was another bullet you mentioned. Network at scale, AI scale, silicon and networking and then AI-enabled networking and screen solutions. Silicon, big advantage, Francisco. We were there when you launched the initiative. How's it going? What's the update? Dave and I were speculating that you should be in the chip business, sell to other people, but... Well, well, we are, we kind of are, already are. Right, you're an asset in San Francisco. You went on one to that. Three and a half years ago. How's that going? Pre-COVID, right, it's like two years ago before we all had to go home. Two COVID years. How about long COVID? That's why I say I got long COVID. Right, you guys announced that. We announced it. Three and a half years ago, it's going very well. We certainly have a number of wins, predominantly amongst the hyperscaler space. Because really, when you think about it, if we take a piece of silicon, we put it on a system, we put software on it, we ship it. Happy to have a business that way, but if customers want to go and take the silicon and put it on their own system, they build themselves, they're kind of shifting the OPEX burden from us to them. And you have to be able to amortize that cost over in number of devices. And so hyperscalers have a lot of devices. And so they're very willing to go down that path of billets. Some of them are, not all of them. Now that flexibility of take it your way, how would you like your burger has gone over very well with customers. Even customers who don't want to go down that path, the fact that they know that they can buy silicon from us, systems with no software, software with no systems, put it on a white box, it makes us very open. And I think that that three and a half years ago is very good. The one data point I will share with you, we launched one ship three and a half years ago. And traditionally what we've seen in this market is people are putting out a new chip every 18 to 24 months. Some even slower than that. In the last three and a half years, we now have 14 chips in that family. So our pace of innovation is unparalleled in the industry and we continue to push very aggressively to get higher, faster speeds and actually go down market with those chips as well. All the actions of the physics right now, the physics is key, the physical layer, my say the old OSI model, Dave, remember the old OSI model? Nailed the physical, kind of getting into the stack there. You guys do great work, props to you guys. Final question I have, and then we got to maybe look at breaking and maybe get one more for you in. As you said, if it's connected, it's protected. Okay, nice tagline. Talk about security. Is Cisco a security company? I mean I tweeted this morning, it's not about point solutions because that's kind of what you were kind of saying today. Security's embedded, NFL's relying on you. It's a huge market by the category we're briefing today. Security, big part of Cisco. Explain, take a minute, explain what is the position of the security market that Cisco play in? Is it embedded in other devices? Is it standalone, point solution, platform? Well, the way I like to think about it is this. So we are absolutely a security player. Security is, as we have said and Chuck has said, is our highest priority. Like we need to make sure that our customers feel like we are doing and building phenomenal products in security space. And I think G2 and Tom and Raj and team have been doing an absolute phenomenal job. Now there's different markets and I don't want to overcomplicate everything. So if somebody wants a firewall, we want to win that business. We want to have the absolute best and we do have the absolute best firewall in the market. If somebody wants an outcome, hey, help me with my branch. You need connectivity. You need security. You might need cameras. You need SD-WAN. You need all these things. Hey, we want to help you with that outcome. But if somebody wants a switch, we're going to have the best darn switch in the entire world and we want you to buy our switch. So we don't want to overcomplicate if somebody's just going to be a switch and then we're just like, well let's talk about what outcomes you want with that switch. It's like if they want a switch, we're going to sell you a switch. But if they're going, hey, I have a challenge here with my branch office or I'm trying to re-envision what this campus looks like on Fifth Avenue, New York. Hey, let's reimagine together and we have all the pieces that you need to have a delightful outcome for the users inside of your business. Final word, bring us home with the interview. What do you want people to walk away with this week? What's your, from your perspective, your keynote? What's the bumper sticker? What's the core message? All right, so a few. Ready? I got about 20 minutes. All right, so. 30 seconds. Unified experiences are critically important. Security and networking are coming together like no one else on the planet can do. And top of that, you've got observability where we have more data than anyone else on the planet and we were able to help you understand what's happening across your infrastructure better than anyone else. And if you haven't taken a look at our full stack of observability with ThousandEyes and App, do you truly, truly, truly are missing out? ThousandEyes looking good. I love the data story. John, thanks for coming on theCUBE. Appreciate your time. I know you're super busy with customers. It's great to be here. Thank you so much. That was a tough thank you month. Okay, it's theCUBE coverage. Day two here at Cisco Live in Las Vegas. I'm John Furrier with Dave Vellante. Analyzing all the action, bringing it to you no matter what. This is our pop-up queue. We're happy to be here. We'll be right back with more after this short break.