 Welcome to Future Cloud made possible by Cisco. My name is Dave Vellante and I'm your host. You know, the cloud is evolving. Like the universe, it's expanding at an accelerated pace. No longer is the cloud just a remote set of services, you know, somewhere up there. No, the cloud, it's extending to on-premises. Data centers are reaching into the cloud through adjacent locations. Clouds are being connected together to each other and eventually they're going to stretch to the edge and the far edge. Workloads, location, latency, local laws and economics will define the value customers can extract from this new cloud model which unifies the operating experience independent of location. Cloud is moving rapidly from a spare capacity slash infrastructure resource to a platform for application innovation. Now the challenge is how to make this new cloud simple, secure, agile and programmable. Oh, and it has to be cloud agnostic. Now the real opportunity for customers is to tap into a layer across clouds and data centers that abstracts the underlying complexity of the respective clouds and locations. And it's got to accommodate both mission critical workloads as well as general purpose applications across a spectrum cost effectively. Enabling simplicity with minimal labor cost requires infrastructure, i.e. hardware, software, tooling, machine intelligence, AI and partnerships within an ecosystem. And it's got to accommodate a variety of application deployment models like serverless and containers and support for traditional work on VMs. By the way, it also requires a roadmap that will take us well into the next decade because the next 10 years, they will not be like the last. So why are we here? Well, theCUBE is covering Cisco's announcements today that connect next generation compute, shared memory, intelligent networking and storage resource pools, bringing automation, visibility, application assurance and security to this new decentralized cloud. Now of course, in today's world, you wouldn't be considered modern without supporting containers, AI and operational tooling that is demanded by forward-thinking practitioners. So sit back and enjoy theCUBE's special coverage of Cisco's Future Cloud. From around the globe, it's theCUBE, presenting Future Cloud, one event, a world of opportunities, brought to you by Cisco. We're here with Vjoy Pandey, a VP of emerging tech and incubations at Cisco. Vjoy, good to see you, welcome. Good to see you as well. Thank you Dave and pleasure to be here. So in 2020, we kind of had to redefine the notion of agility when it came to digital business or organizations, they had to rethink their concept of agility and business resilience. What are you seeing in terms of how companies are thinking about their operations in this sort of new abnormal context? Yeah, I think that's a great question. I think what we're seeing is that pretty much the application is the center of the universe. And if you think about it, the application is actually driving brand recognition and the brand experience and the brand value. So the example I like to give is, think about a banking app, pre-COVID, that did everything that you would expect it to do. But if you wanted to withdraw cash from your bank, you would actually have to go to the ATM and punch in some numbers and then look at your screen and go through a process and then finally withdraw cash. Think about what that would do in a post-pandemic era where people are trying to go contactless. And so in a situation like this, the digitization efforts that all of these companies are going through and the modernization and the automation is what is driving brand recognition, brand trust and brand experience. Yeah, so I was gonna ask you when I heard you say that, I was gonna say, well, but hasn't it always been about the application, but it's different now, isn't it? So I wonder if you could talk more about how the application experience is changing. Yes, as a result of this new digital mandate, but how should organizations think about optimizing those experiences in this new world? Absolutely, and I think, yes, it's always been about the application, but it's becoming the center of the universe right now because all interactions with customers and consumers and even businesses are happening through that application. So if the application is unreliable or if the application is not available, is untrusted, insecure, there's a problem. There's a problem with the brand, with the company and the trust that consumers and customers have with our company. So if you think about an application developer, the weight he or she is carrying on their shoulders is tremendous because you're thinking about rolling features quickly to be competitive. That's the only way to be competitive in this world. You need to think about availability and resiliency like you pointed out and the experience. You need to think about security and trust. Am I as a customer or a consumer willing to put my data in that application? So velocity, availability, security and trust and all of that depends on the developer. So the experience, the security, the trust, the feature velocity is what is driving the brand experience now. So are those two tensions, let's say agility and trust, zero trust used to be a buzzword now. It's a mandate, but are those two vectors counterposed? Can they be merged into one and not affect each other? Does the question make sense? Right, security usually handcuffs my speed, but how do you address that? Yeah, that's a good question. I think if you think about it today, that's the way things are. And if you think about this developer, all they want to do is run fast because they want to build those features out and they're going to pick and choose the APIs and services that matter to them and build out their app. And they want the complexities of the infrastructure and security and trust to be handled by somebody else. It's not that they don't care about it, but they want that abstraction so that it's handled by somebody else. And typically within our organization, we've seen in the past where this friction between NetOps, SecOps, ITOps and the cloud platform teams and the developer on one side and these frictions and these meetings and toil actually take a toll on the developer. And that's why companies and apps and developers are not as agile as they would like to be. So I think, but it doesn't have to be that way. So I think if there was something that would allow a developer to pick and choose, discover the APIs that they would like to use, connect those APIs in a very simple manner and then be able to scale them out and be able to secure them. And in fact, not just secure them during the runtime when it's deployed, but right off the bat when they fire up that IDE and start developing the application, wouldn't that be nice? And as you do that, there is a smooth transition between that discovery, connectivity and ease of consumption and security with the ITOps, NetOps, SecOps teams and CISOs to ensure that they're not doing something that the organization won't allow them to do in a very seamless manner. I want to come back and talk about security, but I want to add another complexity before we do that. So for a lot of organizations, I mean, the public cloud became a staple of keeping the lights on during the pandemic, but it brings new complexities and differences in terms of latency, security, which I want to come back to, deployment models, you know, et cetera. So what are the some of the specific networking challenges that you've seen with the cloud native architectures? How are you addressing those? Yeah, in fact, if you think about cloud to me, that is a different way of saying a distributed system. And if you think about a distributed system, what is at the center of that distributed system is the network. So my favorite comment here is that the network is the runtime for all distributed systems and modern applications. And that is true because if you think about where things are today, like you said, there's cloud assets that a developer might use. In the banking example that I gave earlier, I mean, if you want to build a contactless app so that you get verified, a customer gets verified on the app, they walk over to the ATM and they withdraw cash without touching that ATM. In that kind of an example, you are touching the mobile iOS, let's say iOS APIs, you're touching cloud APIs where the backend might sit, you're touching on-prem APIs, maybe it's an Oracle database or a mainframe, even where transactional data exists. You're touching branch APIs where the ATM actually exists and there needs to be consistency when you withdraw cash and you're carrying all of this and in fact, there might be customer data sitting in Salesforce somewhere. So it's cloud APIs, it's on-prem, it's branch, it's SaaS, it's mobile and you need to bring all of these things together and over time you'll see more and more of these APIs coming from various SaaS providers. So it's not just cloud providers, but SaaS providers that the developer has to use and so this complexity is very, very real and this complexity is across the wide open internet. So the application is built across this wide open internet. So the problems of discoverability, the problems of being able to simply connect these APIs and manage the data flow across these APIs, the problems of consistency of policy and consumption because all of these APIs have their own nuances and what they mean, what the arguments mean and what the API actually means. How do you make it consistent and easy for the developer? That is the networking problem and that is the problem of building out this network, making traffic engineering easy, making policy easy, making scale out, scale down easy, all of that are networking problems and so we are solving those problems at Cisco. Yeah, the internet is a new private network but it's not always so private. So I want to come back to security. You know, I often say that the security model of building a moat, you dig the moat, you get the hardened castle, that's just outdated now. The queen is left her castle, I always say and it's dangerous out there. And the point is, and you touched on this, it's a huge decentralized system and with distributed apps and data, that notion of perimeter security, it's just no longer valid. So I wonder if you could talk more about how you're thinking about this problem and you definitely addressed some of that in your earlier comments, but what are you specifically doing to address this and how do you see it evolving? Yeah, I mean, that's a very important point. I mean, I think if you think about again, the wide open internet being the runtime for all modern applications, what is perimeter security in this new world? I mean, it's to me, it boils down to securing an API because again, going with that running example of this contactless cash withdrawal feature for a bank, the API, wherever it sits, on-prem branch, SaaS, cloud, iOS, Android, doesn't matter. That API is your new security perimeter and the data object that it's trying to access is also the new security perimeter. So if you can secure API to API communication and API to data object communication, you should be good. So that is the new frontier, but guess what? Software is buggy. Everybody's software, I'm not saying Cisco software, everybody's software is buggy. Software is buggy, humans are not reliable and so things mature, things change, things evolve over time, so there needs to be defense in depth. So you need to secure at the API layer, add the data object layer, but you also need to secure at every layer below it so that you have good defense in depth if any layer in between is not working out properly. So for us, that means ensuring API to API communication, not just during runtime, when the app has been deployed and is running, but during deployment and also during the development lifecycle. So as soon as the developer launches an IDE, they should be able to figure out that this API security use is reputable. It is compliant to my organization's needs because it is hosted, let's say from Germany and my organization wants APIs to be used only if they're being hosted out of Germany. So compliance needs and security needs and reputation, is it available all the time? Is it secure? And being able to provide that feedback all the time between the security teams and the developer teams in a very seamless real time manner is again, that's something that we're trying to solve through some of the services that we're trying to produce inside of Cisco. Yeah, I mean, that layered approach that you're talking about is critical because every layer has some vulnerability and so you've got to protect that with some depth. In terms of thinking about security, how should we think about where Cisco's primary value add is? I mean, there's parts of the interview, you guys have great security business, it's a growing business. Is it your intention to add value across the entire value chain? I mean, obviously you can't do everything, so you got a partner, but how should we think about Cisco's role over the next, I'm thinking longer term over the next decade? Yeah, I mean, I think so, we do come in with good strength from the runtime side of the house. So if you think about the security aspects that we have in play today, there's a significant set of assets that we have around user security, around, with Duo and Passwordless, we have significant assets in runtime security. I mean, the entire portfolio that Cisco brings to the table is around runtime security, the secure X aspects around posture and policy that we bring to the table. And as you see Cisco evolve over time, you will see us shifting left. I mean, I know it's an overused term, but that is where security is moving towards. And so that is where API security and data security are moving towards. So learning what we have during runtime, because again, runtime is where you learn what's available. And that's where you can apply all of the ML and AI models to figure out what works, what doesn't. Taking those learnings, taking those catalogs, taking that reputation database and moving it into the deployment and development lifecycle and making sure that that's part of that entire dev to deploy to runtime chain is what you will see Cisco do over time. That's fantastic, phenomenal perspectives we get. Vijay, thanks for coming on theCUBE. Great to have you and look forward to having you again. Absolutely, thank you. In a moment, we'll talk hybrid cloud applications, operations and potential gaps that need to be addressed with Kostub Das and Vijay Venagopal. You're watching theCUBE, the global leader in high tech coverage. Your cloud, it isn't just a cloud. It's everything flowing through us. It's alive, connecting users, applications, data and devices. And whether it's cloud native, hybrid or multi-cloud, it's more distributed than ever. One company takes you inside, giving you the visibility and the insight you need to take action. One company has the vision to understand it all, the experience to securely connect it all. On any platform, in any environment. So you can work wherever work takes you. In a cloud-first world, between your cloud and being cloud smart, there's a bridge, Cisco, the bridge to possible. Okay, we're here with Kostub Das, who's the Senior Vice President, General Manager of Cloud and Compute at Cisco and Vijay Venagopal, who is the Senior Director for Product Management for Cloud Compute at Cisco. KD, Vijay, good to see you guys, welcome. Great to see you, Dave. Good to be here. KD, let's talk about cloud. You and I, last time we were face to face was in Barcelona, we loved talking about cloud. And I always say to people, look, Cisco's not a hyperscaler, but the big public cloud players, they're like giving you a gift. They spent almost, actually over $100 billion last year on CapEx, the big four. So you can build on that infrastructure. Cisco's all about hybrid cloud. So help us understand the strategy there, maybe how you can leverage that build out. And importantly, what are customers telling you they want out of hybrid cloud? Yeah, no, that's a perfect question to start with, Dave. So yes, so the hyperscalers have invested heavily building out their assets. There's a great lot of innovation coming from that space. There's also a great innovation, sort of innovation coming from open source. And that's another source of a gift, in fact, to the IT community. But when I look at my customers, they're saying, well, how do I, in the context of my business, implement a strategy that takes into consideration everything that I have to manage in terms of my contemporary workloads, in terms of my legacy, in terms of everything my developer community wants to do on DevOps and really harness that innovation that's built in the public cloud, that built an open source, that built internally to me. And that naturally leads them down the path of a hybrid cloud strategy. And Cisco's mission is to provide for that imperative the simplest, more powerful platform to deliver hybrid cloud. And that platform is inner-site. We've been investing in inner-site, it's a SaaS service. Inner-site delivers to them that bridge between their states of today, their workloads of today, the need for them to be guardians of enterprise-grade resiliency with the agility that's needed for the future, the embracing of cloud native, of new paradigms of DevOps models, the embracing of innovation coming from public cloud and an open source. And bridging those two is what inner-site has been doing. That's kind of the crux of our strategy. Of course, we have the entire portfolio behind it to support any version of that, whether that is on-prem in the cloud, hybrid cloud, multi-cloud and so forth. But if I understand it correctly from what I heard earlier today, the inner-site is really a linchpin of that strategy. Is it not? It really is. And may take a second to really familiarize those who don't know inner-site with what it is. We started building this platform quite a few years back and we built it ground up to be a immensely scalable, SaaS, super simple hybrid cloud platform. And it's a platform that provides a slew of services inherently. And then on top of that, there are suites of services. There are suites of services that are tied to infrastructure automation, Cisco as well as Cisco partners. There are suites of services that have nothing to do with Cisco products from a hardware perspective and it's got to do with more cloud orchestration and cloud natives. And inner-site and its suite of services continue to kind of increase in pace and velocity of delivery. Just over the last two quarters, we've announced a whole number of things. We'll go a little bit deeper into some of those, but they span everything from infrastructure automation to Kubernetes and delivering Kubernetes as a service to workload optimization and having visibility into your cloud estate, how much it's costing into your on-premise state, into your workloads and how they're performing. It's got integrations with other tooling with both Cisco, AppD, as well as non-Sysco assets. And then it's got a whole slew of capabilities around orchestration because at the end of the day, the job of IT is to deliver something that works and works at scale that you can monitor and make sure is resilient. And that includes a workflow, an ability to say, do this, then do this and do this or it includes other ways of automation like infrastructure code and so forth. So it includes self-service that span that, but we're inside the world's simplest hybrid cloud platform, rapidly evolving, rapidly delivering new services and we've talked about some more of those there. Great, thank you, KD. Vijay, let's bring you into the discussion. You guys recently made an announcement with HashiCorp, I was stoked because even though it seemed like a long time ago, pre-COVID, in my predictions post, I said Hashi was a name to watch, our data partners, ETR, you look at the survey data and they really have become mainstream. Particularly, we think very important in the whole multi-cloud discussion and as well, they're attractive to customers. They have open source offerings, you can very easily experiment, smaller organizations can take advantage, but if you want to upgrade to enterprise features like clustering or whatever, you could plug right in, not a big complicated migration. So a very, very compelling story there. Why is this partnership important to Cisco's customers? Absolutely, Dave, and spot on every single thing that you said. Let me just start by paraphrasing what our mission statement is in the cloud and compute group. Our mission statement is to enable a cloud operating model for hybrid cloud and what we mean by that is the ability to have extreme amounts of automation, orchestration and observability across your hybrid cloud IT operations. Now, so developers and applications team get a great amount of agility in public clouds and we're on a mission to bring that kind of agility and automation to the private cloud and to the data centers. And InterSight is a key platform and linchpin to enable that kind of operations, cloud-like operations in the private clouds. And key, as you rightly said, HashiCorp is the, they were the inventors of a concept of infrastructure as code and in Terraform, they have the world's number one infrastructure as code platform. So it became a natural partnership for Cisco to enter into a technology partnership with HashiCorp to integrate InterSight with HashiCorp's Terraform to bring the benefits of infrastructure as code to the two hybrid cloud operations. And we've entered into a very tight integration and a partnership where we allow developers, DevOps teams and infrastructure administrators to allow the use of infrastructure as code in a SaaS-delivered manner for both public and private cloud. So it's a very unique partnership and a unique integration that allows the benefits of cloud-managed IAC to be delivered to hybrid cloud operations. And we've been very happy and proud to be partnering with HashiCorp on that. Yeah, Terraform gets very high marks from customers, a lot of value there. The InterSight integration adds to that value. Let's stay on cloud native for a minute. We all talk about cloud native. KD was sort of mentioning before, you got the core apps. You want to protect those, make sure their enterprise create, but they got to be cool as well for developers. You're connecting to other apps in the cloud or wherever. How are you guys thinking about this cloud native trend? What are the moves you making in this regard? I mean, cloud native is one of the paramount ID trends of today. And we're seeing massive amounts of adoption of cloud native architectures in all modern applications. Now, cloud native has become synonymous with Kubernetes these days. And Kubernetes has emerged as a de facto cloud native platform for modern cloud native app development. Now, what Cisco has done is we have created a brand new SaaS-delivered Kubernetes service that is integrated with InterSight. We call it the InterSight Kubernetes Service, or IKS, and this just geared a little over one month ago. Now, what InterSight Kubernetes Service does is it delivers a cloud-managed and cloud-delivered Kubernetes service that can be deployed on any supportive target infrastructure. It could be a Cisco infrastructure, it could be a third-party infrastructure, or it could even be public cloud. But think of it as Kubernetes anywhere delivered as SaaS managed from InterSight. So very powerful capability that we've just released into InterSight to enable the power of Kubernetes and cloud native to be used anywhere. But today we made a very important announcement because we have today announced the brand new Cisco Service Mesh Manager, the Cisco Service Mesh Manager, which is available as an extension to IKS or to InterSight. Basically, we see service meshes as being the future of networking, right? In the past, we had layer two networking and layer three networking. And now with service meshes, application networking and layer seven networking is the next frontier of networking. But you need to think about networking for the application age very differently, how it is managed, how it is deployed. It needs to be very developer-friendly and developer-centric. And so what we have done is we've built out an application networking strategy and built out the Service Mesh Manager as a very simple way to deliver application networking to the consumers like developers and application teams. This is built on an acquisition that Cisco made recently of Bonsai Cloud and we've taken the assets of Bonsai Cloud and delivered the Cisco Service Mesh Manager as an extension to IKS that brings the promise of future networking and modern networking to application and developer teams. Got it. Thank you, Vijay. And so, Katie, let's wrap this up. I mean, there was a lot in this announcement today. A lot of themes around openness, heterogeneity and a lot of functionality and value. Give us your final thoughts. Absolutely. So a couple of things to close on. First of all, InterSight is the simplest, most powerful hybrid cloud platform out there. It enables that cloud operating model that Vijay talked about, but it enables that across cloud. So it's sass. It's relatively easy to get into it and give it a spin so that I would highly encourage anybody who's not familiar with it to try it out and anybody who is familiar with it to look at it again because there are probably services in there that you didn't notice or didn't know last time you looked at it because we're moving so fast. So that's the first thing. The second thing I'll close with is we've been talking about this bridge that's kind of bridging your on-prem, your open source, your cloud estates. And it's so important to make that mental leap because in past generation, we used to talk about integrating technologies together. And then with public cloud, we started talking about move to public cloud. But it's really how do we integrate? How do we integrate all of that innovation that's coming from the hyperscalers, everything they're doing to innovate super fast? All of that innovation that's coming from open source, all of that innovation that's coming from companies around the world, including Cisco, how do we integrate that to deliver an outcome? Because at the end of the day, if you're a cloud ops team, if you're an IT ops team, your job is to deliver an outcome and our mission is to make it super simple for you to do that. So that's the mission we're on and we're hoping that everybody that's excited as we are about how simple we made that. Great, thank you a lot in this announcement today. Appreciate you guys coming back on and help us unpack some of the detail. Thanks so much, great having you. Thank you, Dave. In a moment, we're gonna come back and talk about disruptive technologies and futures in the age of hybrid cloud with Vikas Ratna and James Leach. You're watching theCUBE, the global leader in high tech coverage. What if your server box wasn't a box at all? What if it could do anything? Run anything, be any box you need with massive scale, precision and intelligence. Managed and optimized from the cloud, integrated with all your clouds, private, public, or hybrid. So you can build whatever you need today and tomorrow. The potential of this box is unlimited, unstoppable, unseen ever before. Unbox the future with Cisco UCSX series, powered by InterSight. Cisco, the bridge to possible. We're here with Vikas Ratna, who's the director of product management for ECS at Cisco and James Leach is the director of business development for UCS at Cisco as well. We're gonna talk about computing in the age of hybrid cloud. Welcome gentlemen, great to see you. Thank you. Thank you. Vikas, let's start with you and talk a little bit about computing architectures. We know that they're evolving, they're supporting new data intensive and other workloads, especially these high performance workload requirements. What's Cisco's point of view on all this? I mean, specifically, I'm interested in your thoughts on fabrics, I mean, it's kind of your wheelhouse. You've got accelerators. What are the workloads that are driving these evolving technologies? And how is it impacting customers? What are you seeing? Sure, Dave. First of all, very excited to be here today. You're absolutely right. The pace of innovation in foundational platform ingredients have just been phenomenal in recent years. The fabric, the accelerators, the drives, the processing power, the core density, all have been evolving at just an amazing pace. And the pace will only pick up further. But ultimately, it is all about applications and the way applications leverage those innovations. And we do see applications evolving quite rapidly. The new classes of applications are evolving to absorb those innovations and deliver much better business values. Very, very exciting time step. But talking about the impact on the customer, well, these innovations have helped them very positively. We do see significant challenges in the data center with the point product-based approach of delivering these platform innovations to the applications. What has happened is these innovations today are being packaged as point products to meet the needs of a specific application. And as you know, the different applications have no different needs. Some applications need more GPUs. Others need more memory. Yet others need more cores. Some need different kind of fabrics. As a result, if you walk into a data center today, it is very common to see many different point products in the data center. This creates a manageability challenge. Imagine the aspect of managing, several different form factors, one you, two you, purpose-built servers, the variety of a blade form factor. This reminds me of the situation we had before smartphones arrived. You remember the days when we used to have a GPS device for navigation system, a cool music device for listening to the music, a phone device for making a call, camera for taking the photos. And we were all excited about it. It's when smartphones arrived that we realized all those cool innovations could be delivered in a much simpler, much convenient and easy to consume through one device and that could completely transform our experience. So we see the customers who are benefiting from these innovations to have a way to consume those things in a much more simplistic way than they are able to do today. And I like the, look, it's always been about the applications, but to your point, the applications are now moving at a much faster pace. The customer experience is, expectation is way escalated. And when you combine all these, I love your analogy there, because when you combine all these capabilities, it allows us to develop new applications, new capabilities, new customer experiences. So that's, I always say the next 10 years, they ain't going to be like the last. And James. Absolutely. Public cloud obviously is heavily influencing compute design and customer operating models. You know, it's funny, when the public cloud first hit the market, everyone, we were swooning about, oh, low cost standard off the shelf servers, you know, in storage devices, but it quickly became obvious that customers needed more. So I wonder if you could comment on this. How are the trends that we've seen from the hyperscalers, how are they filtering into on-prem infrastructure? And maybe, you know, maybe there's some differences there as well that you could address. Absolutely. So, you know, I'd say first of all, quite frankly, you know, public cloud has completely changed the expectations of how our customers want to consume compute, right? So customers, especially in a public cloud environment, they've gotten used to or, you know, come to accept that they should consume from the application out, right? They want a very application focused view, a services focused view of the world. They don't want to think about infrastructure, right? They want to think about their application. They want to move outward, right? So this means that the infrastructure basically has to meet the application where it lives. So what that means for us is that, you know, we're taking a different approach. We've decided that, you know, we're not going to chase this, you know, single pane of glass view of the world, which, you know, frankly, our customers don't want. They don't want a single pane of glass. What they want is a single operating model. They want an operating model that's similar to what they can get with the public cloud, but they want it across all of their cloud options. They want it across private cloud, across hybrid cloud options as well. So what that means is they don't want to just consume infrastructure services. They want all of their cloud services from this operating model. So that means that they may want to consume infrastructure services for automation and orchestration, but they also need Kubernetes services. They also need virtualization services. They may need Terraform, workload optimization. All of these services have to be available from within the operating model, a consistent operating model, right? So it doesn't matter whether you're talking about private cloud, hybrid cloud, anywhere, where the application lives, doesn't matter. What matters is that we have a consistent model that we think about it from the application out. And frankly, I'd say, you know, this has been the stumbling block for private cloud. Private cloud is hard, right? This is why it hasn't been really solved yet. This is why we had to take a brand new approach. And frankly, it's why we're super excited about X-Series and InterSight as that, that, you know, operating model that fits the hybrid cloud better than anything else we've seen. This is a cube first. First time's a technology vendor has ever said that it's not about a single pane of glass because I've been hearing for decades, we're going to deliver a single pane of glass and it's going to be seamless and it never happens. It's like a single version of the truth. It's aspirational and it's just not reality. So can we stay in the X-Series for a minute, James? Maybe in this context. But in the launch that we saw today was like a fire hose of announcement. So how does the X-Series fit into the strategy with InterSight and hybrid cloud and this operating model that you're talking about? Right, so I think it goes hand in hand, right? The two pieces go together very well. So we have this idea of a single operating model that is definitely something that our customers demand, right? It's what we have to have. But at the same time, we need to solve the problems Vikas was talking about before. We need a single infrastructure to go along with that single operating model. So no longer do we need to have silos within the infrastructure that give us different operating models or different sets of benefits when you want infrastructure that can kind of do all of those configurations, all those applications. And then the operating model is very important because that's where we abstract the complexity that could come with just throwing all that technology at the infrastructure. So the way that we think about it is the data center is not centered, right? It's no longer centered. Applications live everywhere. Infrastructure lives everywhere. And we need to have that consistent operating model but we need to do things within the infrastructure as well to take full advantage, right? So we want all the SaaS benefits of a CICD model of the intersite can bring. We want all that proactive recommendation engine with the power of AI behind it. We want the connected support experience. We want all of that, but we want to do it across a single infrastructure. And we think that that's how they tie together. That's why one or the other doesn't really solve the problem. But both together, that's why we're here. That's why we're super excited. So Vikas, I make you laugh a little bit. When I was an analyst at IDC, I was a bit deep into infrastructure and then when I left, I was doing, I was working with application development heads. And like you said, infrastructure was just a roadblock. But so the target chic is when Cisco announced UCS a decade ago, I totally missed it. I didn't understand it. I thought it was Cisco getting into the traditional server business. And it wasn't until I dug in that I realized that your vision was really to transform infrastructure deployment and management and change them all. I was like, okay, I got that wrong. But so let's talk about the ecosystem and the joint development efforts that are going on there. X-Series, how does it fit into this converged infrastructure business that you've built and grown with partners? You got storage partners like NetApp and Pure. You got ISV partners in the ecosystem. We see Cohesity. It's been a while since we hung out with all these companies at Cisco Live, hopefully next year. But tell us what's happening in that regard. No, absolutely. I'm looking forward to seeing you in the Cisco Live next year, Dave. Absolutely. You brought up a very good point. UCS is about the ecosystem that it brings together. It's about making our customers bring up the entire infrastructure from the core foundational hardware all the way to the application level so that they can go off and running pretty quick. The converged infrastructure has been one of the cornerstones of the strategy as you pointed out in the last decade. And I'm very glad to share that converged infrastructure continues to be a very popular architecture for several enterprise applications even today. In fact, it is the preferred architecture for mission-critical applications where performance, resiliency, latency, are the critical requirements. They are almost a de facto standards for large-scale deployments of virtualization, business-critical databases, and so forth. With X-Series, with our partnerships, with our storage partners, those architectures will absolutely continue and will get better. But in addition, it's a hybrid cloud world. So we are now bringing in the benefits of converged infrastructure to the world of hybrid cloud. We'll be supporting the hybrid cloud applications now with this CI infrastructure that we have built together with our strong partnership with the storage partners to deliver the same benefits to the new AIS applications as well. Yeah, and that's what customers want. They want that cloud operating model, right? Go ahead, please. I was just going to say, the CI model will continue to thrive. It will transition, it will expand the use cases now for the newer use cases that we're beginning to say, Dave, absolutely. All right, great. Thank you for that. And James, as I said earlier today, we heard this huge announcement. A lot of parts to it. And we heard KD talk about this initiative as it's really computing built for the next decade. I mean, I like that because it shows some vision and that you've got a roadmap that you've thought through the coming changes in workloads and infrastructure management and some of the technology that you can take advantage of beyond just one or two product cycles. So, but I want to understand what you've done here specifically that you feel differentiates you from other competitive architectures in the industry. Sure, that's a great question, number one. Number two, I'm frankly a little bit concerned at times for customers in general, for our customers, customers in general, because if you look at what's in the market, right? These rinse and repeat systems that were effectively just rehashes of the same old design, right? That we've seen since before 2009 when we brought UCS to market. These are what we're seeing over and over and over again. That's not really gonna work anymore, frankly. And I think that people are getting lulled into a false sense of security by seeing those things continually put in the market. We'd rethought this from the ground up because frankly, future-proofing starts now, right? If you're not doing it right today, future-proofing isn't even on your radar because you're not even today-proofed. So, we've rethought the entire chassis, the entire architecture from the ground up. Okay, if you look at other vendors, if you look at other solutions in the market, what you'll see is things like, management inside the chassis, that's a great example. Daisy chaining them together, who needs that? Who wants that? That kind of complexity is, first of all, it's ridiculous. Second of all, if you want to manage across clouds, you have to do it from the cloud, right? It's just common sense. You have to move management where it can have the scale and the scope that it needs to impact your entire domain, your world, which is much larger now than it was before. We're talking about true hybrid cloud here, right? So, we had to solve certain problems that existed in the traditional architecture. I can't tell you how many times I heard, talk about the midplane is a great example. The midplane in a chassis is a limiting factor. It limits us on how much we can connect or how much bandwidth we have available to the chassis. It limits us on airflow and other things. So, how do you solve that problem? Simple, just get rid of it. We took it out, right? It's no longer a problem. We designed an architecture that doesn't need it. It doesn't rely on it. No forklift upgrades. So, as we start moving down the path of needing liquid cooling, or maybe we need to take advantage of some new high performance, low latency fabrics, we can do that with almost no problem at all, right? So, we don't have any forklift upgrades. Park your forklift on the side. You won't need it anymore because you can upgrade granularly. You can move along as technologies come into existence that maybe don't even exist today. They may not even be on our radar today to take advantage of. But I like to think of these technologies, they're really important to our customers. These are, we can call them disruptive technologies. The reality is that we don't want to disrupt our customers with these technologies. We want to give them these technologies so they can go out and be disruptive themselves, right? And this is the way that we've designed this from the ground up to be easy to consume and to take advantage of what we know about today and what's coming in the future that we may not even know about. So, we think this is a way to give our customers that ultimate capability, flexibility and future proofing. I like that phrase true hybrid cloud. It's one that we've used for years. But to me, this is all about that horizontal infrastructure that can support that vision of what true hybrid cloud is. You can support the mission critical applications. You can develop on the system and you can support a variety of workloads locked into one narrow stovepipe. And that does have legs. Vikas and James, thanks so much for coming on the program, great to see you. Thank you. I appreciate the time. When we return shortly, Thomas Scheiba who leads Cisco's data center group will be here. And Thomas has some thoughts about the transformation of networking IT teams. You don't want to miss what he has to say. You're watching theCUBE, the global leader in high tech company. Okay, we're here with Thomas Scheiba who's the vice president of product management AKA VP of all things, data center, networking, SDN cloud you name it in that category. Welcome Thomas, good to see you again. Hey, Sam, yes, thanks for having me on. Yeah, it's our pleasure. Okay, let's get right into observability. When you think about observability, visibility, infrastructure monitoring, problem resolution across the network, how does cloud change thing? In other words, what are the challenges that networking teams are currently facing as they're moving to the cloud and trying to implement hybrid cloud? Yeah, visibility as always is very, very important. And it's quite frankly, it's not just the network team it's actually the application team too, right? And as you pointed out, the underlying impetus to what's going on here is the data center is where the data is. And I think we said as a couple of years back and really what happens, the applications are going to be deployed in different locations, right? Whether it's in a public cloud, whether it's on-prem and they're built differently, right? They're built as microservices that might actually be distributed as well at the same application. And so what that really means is you need as an operator as well as actually a user a better visibility where are my pieces? And you need to be able to correlate between where the app is and what the underlying network is that has some place in these different locations. So you have actually a good knowledge why the app is running so fantastic or sometimes not. So I think that's really the problem statement what we're trying to go after was observability. Okay, and let's double click on that. So a lot of customers tell me that you got to stare at log files until your eyes bleed and you got to bring in guys with lab coats who have PhDs to figure all this stuff out. So you just described it's getting more complex but at the same time you have to simplify things. So how are you doing that? Correct, so what we basically have done is we have this fantastic product that is called Thousand Eyes. And so what this does is basically has the name which I think is a fantastic name. You have these sensors everywhere and you can have a good correlation on links between if I run from a site to a site, from a site to a cloud, from a cloud to a cloud and you basically can measure what is the performance of these links. And so what we're doing here is we're actually extending the footprint of the Thousand Eyes agent, right? Instead of just having an inversion machine of clouds we are now embedding them with the Cisco network devices, right? We announced this with the Catalyst 9000 and we're extending this now to our 8000 Catalyst product line for the SD-WAN products as well as to the data center products, the nexus line. And so what you see is as you're half a zing you have Thousand Eyes, you get a million insights and you get a billion dollar of improvements for how your applications run. And this is really the power of tying together the footprint of what a network is with the visibility what is going on. So you actually know the application behavior that is attached to this network. I see, so okay, so as the cloud evolves it expands, it connects. You're actually enabling Thousand Eyes to go further not just confined within a single data center location but out to the network across clouds, et cetera. Correct, wherever the network is you're going to have a Thousand Eyes sensor and you can bring this together and you can quite frankly pick. If you want to say, hey, I have my application in public cloud provider A domain one and I have another domain two I can do monitor that link. I can also monitor, I have a user that has a campus location or branch location. I kind of put an agent there and then I can monitor the connectivity from that branch location all the way to the, let's say corporations data center or headquarter or to the cloud and I can have these probes and just to have visibility and saying, hey, if there's a performance, I know where the issue is and then I obviously can use all the other tools that we have to address those. All right, let's talk about the cloud operating model. Everybody tells us that it's really the change in the model that drives big numbers in terms of ROI and I want you to maybe address how you're bringing automation and DevOps to this world of hybrid and specifically, how is Cisco enabling IT organizations to move to a cloud operating model as that cloud definition expands? Yeah, no, that's another interesting topic beyond the observability. So really what we're seeing and this is going on for, I want to say a couple of years now, it's really this transition from operating infrastructure as a networking team, more like a service like what you would expect from a cloud provider, right? It's really around the networking team offering services like a cloud provider does and that's really what the meaning is of cloud operating model, right? Where this is infrastructure running in your own data center where that's linking that infrastructure was whatever runs on the public cloud is operating in like a cloud service. And so we are on this journey for a while. So one of the examples that we have removing some of the control software assets that customers today can deploy on-prem to an instance that they can deploy in a cloud provider and just busy instantiate things there and then just run it that way, right? And so the latest example for this is what we have our identity service engine that is now unlimited availability available on AWS and will become available mid this year both on AWS and Azure as a service. You can just go to Marketplace, you can load it there and now you basically can start running your policy control in a cloud managing your access infrastructure in your data center, in your campus, wherever you want to do it. And so that's just one example of how we see our customers network operations team taking advantage of a cloud operating model and basically applying their tools where they need them and when they need them. So what's the scope of, I hope I'm saying it right, I used to write I think it's called ICE. What's the scope of that? Like for instance, can it affect my or even you know, address simplify my security approach? Absolutely, that's now coming to what is the beauty of the product itself? Yes, what you can do is really is like there's a lot of people talking about else how do I get to a zero trust approach to networking? How do I get to a much more dynamic flexible segmentation in my infrastructure? Again, whether this is only campus X as well as in data center and ICE helps you there. You can use this as a point to define your policies and then interconnect from there, right? In this particular case, if you would instant ICE in a cloud as a software load, you now can connect and say, hey, I want to manage and program my network infrastructure and my data center or my campus going to the respective controller whether it's DNA center for campus or whether it's the ACI policy controller. And so yes, what you get as an effect out of this is a very elegant way to automatically manage in one place what is my policy and then drive the right segmentation in your network infrastructure. Yeah, zero trust. It was pre-pandemic, it was kind of a buzzword. Now it's become a mandate. I wonder if we could talk about, yeah, right? I mean, so I want to talk about cloud native apps. You got all these developers that are working inside organizations. They're maintaining legacy apps. They're connecting their data to systems in the cloud. They're sharing that data and these developers, they're rapidly advancing their skill sets. How is Cisco enabling its infrastructure to support this world of cloud native making infrastructure more responsive and agile for application developers? Yeah, so we're going to the top. We've established visibility. We talked about the operating model, how our network operates actually want to use tools going forward. Now the next step to this is it's not just the operator. How do they actually, where do they want to put these tools? How are they interact with these tools? As well as quite frankly, is how let's say a DevOps team, one application team or a cloud team also wants to take advantage of the programmability of the underlying network. And this is where we're moving into this whole cloud native discussion, right? Which has really two angles. There is the cloud native way, how applications are being built. And then there is the cloud native way, how you interact with infrastructure, right? And so what we have done is we're A, putting in place the unramps between clouds. And then on top of it, we're exposing for all these tools, APIs that can be used and leveraged by standard cloud tools or cloud native tools, right? And one example or two examples we always have. And again, we're on this journey for a while is both Ansible script capabilities that exists from Red Hat as well as Hashi Terraform capabilities that you can orchestrate across infrastructure to drive infrastructure automation. And what really stands behind it is what either the networking operations team wants to do or even the app team, they want to be able to describe the application as a code and then drive automatically or programmatically instantiation of infrastructure needed for that application. And so what you see us doing is providing all these capability as an interface for all our networking tools, right? Whether this is I, is what I just mentioned, whether this is our DCN controllers in the data center, whether these are the controllers in the campus. For all of those, we have cloud native interfaces. So an operator or a DevOps team can actually interact directly with that infrastructure the way they would do today with everything that lives on the cloud or with everything how they build the application. Yeah, this is key. You can't even have the conversation of cloud operating model that includes and comprises on-prem without programmable infrastructure. So that's very important. Last question, Thomas, are customers actually using this? You made the announcement today. Are there any examples of customers out there doing this? We do have a lot of customers out there that are moving down the path and using the Cisco high performance infrastructure also on the compute side, as well as on the Nexus side. One of the customers, and this is an interesting case, is Rakuten. Rakuten is a large tech provider, a mobile 5G operator in Japan and expanding in different countries. And so people sometimes think, oh, cloud, you must be talking about the public cloud provider, the big three or four. But if you look at it, there's a lot of the tech or service providers that actually cloud providers as well and expanding very, very rapidly. And so we're actually very proud to work together with Rakuten and helped in building a high performance data center infrastructure based on 100 gig and actually four on a gig to drive their deployment to it's a 5G mobile cloud infrastructure, which is where the whole world is going and so it's really exciting to see these development and see the power of automation visibility together with the high performance infrastructure becoming a reality and delivering actually services. Yes, some great points you're making there. I mean, yes, you have the big four clouds that are enormous, but then you have a lot of actually quite large clouds, telcos that are either proximate to those clouds or they're in places where those hyperscalers may not have a presence and building out their own infrastructure. So that's a great case study. Thomas, hey, great having you on. Thanks so much for spending some time with us. Yeah, same here, I appreciate it. Thanks a lot. I'd like to thank Cisco and our guests today. VJoy, KD, VJ, Vikas, James and Thomas for all your insights into this evolving world of hybrid cloud. As we said at the top, the next decade will be defined by an entirely new set of rules and it's quite possible things will evolve more quickly because the cloud is maturing and has paved the way for a new operating model where everything is delivered as a service. Automation has become a mandate because we just can't keep throwing IT labor at the problem anymore. And with AI, so much more is possible in terms of driving operational efficiencies, simplicity and support of the workloads that are driving the digital transformation that we talk about all the time. This is Dave Vellante and I hope you've enjoyed today's program. Stay safe, be well, and we'll see you next time.