Cisco Systems, Inc. (NASDAQ:CSCO) Goldman Sachs and Cisco Host a Tech Talk on Cisco’s Enterprise Strategy December 3, 2018 12:00 PM ET
Gloria Lee - IR.
Sachin Gupta - SVP, Enterprise Networking
Rod Hall - Goldman Sachs
Good afternoon ladies and gentlemen and welcome to the Tech Talk with Cisco Systems. At this time all participants have been placed on a listen-only mode and we will open the floor for your questions and comments after the presentation.
It is now my pleasure to turn the floor over to your host Rod Hall. Sir, the floor is yours.
Thanks, Catherine, and good morning everyone. Thanks for dialing in. So we have the pleasure of having Sachin Gupta with us, the SVP of Enterprise Networking at Cisco. I'm looking forward to asking a few questions and then at the end of this we’ll open up and see if anybody on the line has questions as well.
But before we do that we've got a couple of disclosures that need to be read. So Gloria Lee from Investor Relations at Cisco is going to start us off with their Reg FD disclosures and then I've got a couple things to read for Goldman Sachs and then we'll get going. Gloria?
Great, thanks, Rod. As a reminder today's call pertains mainly to Cisco's Enterprise Networking business. No new financial information regarding Cisco's overall performance is intended or implied and this call should not be viewed as an update to the quarter. We may make forward-looking statements, which are subject to risks and uncertainties outlined in detail in our documents filed with the SEC, including our most recent filings of Form 10-K and 10-Q. Actual results may differ from statements made today.
In addition a webcast replay of this call will be available on our company web site under the Investor Relations link.
I'll now turn the call over you, Rod.
Great, thanks, Gloria. So in terms of my disclosures, I just need to mention that Alphabet, Amazon.com, Microsoft and VMware Incorporated are covered by other Goldman Sachs analysts and also that Dell Technologies is not covered by Goldman Sachs at this time.
The other thing that I'll say is we are required to make certain disclosures in public appearances about Goldman Sachs relationships with companies that we discuss. The disclosures relate to investment banking relationships, compensation received or 1% or more ownership. We are prepared to read aloud disclosures for any issuer are upon request. However, these disclosures are available in our most recent reports available to you as clients on our firm portals.
So now with that exciting stuff is over, I guess, I'll get it started. So, Sachin maybe for people that are dialed in, I think, most people know you but maybe you could just give us a quick overview of your responsibilities at Cisco, for people that aren’t as familiar with you and highlight kind of key milestones for the Enterprise Networking business over the last year or so.
Thanks, Rod. And thanks for this I am very happy to be here. So quick introduction, again my name is Sachin Gupta, I have been in Cisco now 21 years and I run Product Management for our Enterprise Networking business. That includes our routing and FDRAM [ph] portfolio with Viptela, our Catalyst Switching portfolio, our wireless portfolio, as well as whatever we're doing with intent-based networking in the enterprise with products like DNA Center and our identity services engine.
And, really when you talk about milestones, this is a journey that we started off a few years ago. When we saw that our customers were seeing a lot more happening in access with mobility, with IoT, lots of users' devices and things connecting and now reaching applications and data that could sit in any cloud. This was driving demand in access and in when that was unprecedented.
At the same time, we saw that networking was still pretty manual, it could be pretty complex. And customers who wanted to deliver the right security and experience end-to-end to any kind of user any kind of thing, to any application across all these domains were running into challenges.
We set out to reinvent the network. I mean, that was a bold statement, but we literally rebuilt the entire stack all the way from the ASIC, to the systems, to the operating system on top, to the controller on top of that. And we introduced this, I can't forget the date, because it was such a big date. On June 20th last year 2017, with intent-based networking we came out with a few critical products there the Catalyst 9000 series, IOS XE, the operating system that was built on custom UADP ASIC, that was inside it as a heart of it.
We came up with DNA Center, our controller sort of the command center for the enterprise. We came up with a segmentation automation technology with software defined access. And we even introduced encrypted traffic analytics. The ability to detect malware encrypted flows without decrypting them.
And since then we've been really busy. We added wireless assurance. How do you detect wireless issues very quickly, root cause them, deliver a rich experience for the mobile users. We expanded the technology to IoT, we extended intent-based networking to IoT. We built out API, so we can extend DNA Center as a platform and integrates with the likes of service now.
And most recently at Partner Summit we extend the Cat9K family and talked about new innovations on our Viptela at DRAM side. So it's been a very busy time, but I can name sort of one milestone, the team has been busy at this for a few years and is really great to see these innovations come out to market and help our customers solve problems.
Q - Rod Hall
Okay. Thanks, Sachin. So what I wanted to maybe kick off with you guys just announced the Catalyst 9200 and 9800 wireless controllers and we know those are very -- they're replacing the very popular Catalyst 9200 products. So I'm just wondering if you could maybe give us some idea of what proportion of total catalyst for sales you expect those two models to represent? I know that 2k at least historically been a majority upward* sales.
Yes. And so, let me just explain a little bit about the Catalyst 9200 and what it represents. So you're right, the Catalyst 9200 is a high volume play.
It's targeting mid-market deployments. So think smaller size, less full time employees, and branch deployments. So you may have simple brunches with a few employees, but you could be a very large enterprise customer.
We expect that the Catalyst 9200 could represent about 25% of the world's Campus access ports, right. So this is something we've talked about in our Partner Summit as well. So just to give you a sheer idea of size of what we're going after and you're exactly right. It is the next generation of the Cat2K product that historically has also served this high volume space.
And so, think again customers with smaller scale deployment and this really doubles the number of customers we can address with the Cat9K family. So allowing us to take that intent-based networking value and bring it to everybody.
Now some of the key attributes here. Obviously, it is a great new switch on its own. And so it makes things like resiliency default. It now has redundant power supplies, there is table stakes. It has double the performance at approximately the same price. It's built on the same custom ASIC, but a mini version of it so UADP, but the real magic here is in software. There is a first switch, the Cat12K family was not software upgradable.
The 9200 is fully software upgradable supports all the three tiers of subscriptions that we’ve announced DNA Essentials, DNA Advantage, DNA Premier. And it brings the strength of IBM down to all of these additional types of deployment. So software defined access, assurance, network as a sensor for threat detection, the new same modular IOS XE operating system with patching and open APIs are also on the 9200. So same topic consistency, same architectural consistency with intent-based networking.
Okay, that's great. And then how about the 9800, do you want to talk a little bit about what is new there? I guess, all the same software features that maybe how does that integrate with wireless portfolio and what other changes come along with a wireless portfolio as a result of the 9800 or are those yet to be announced?
Yes. So 9800 we announced as well, the 9800 is the next generation of our wireless controller. And our wireless controller has been built on our Internet portfolio, based on AirOs operating system for about 15 years. So there's tremendous RF and Wi-Fi innovation that’s gone in now and we've taken all of that innovation and brought it into IOS XE, again the same release trend, same software on the rest of the Cat9K family.
So first of all, you automatically get those advantages, the open net config API, the hot patching support software defined access, encrypted traffic analytics, all of those things come in just because it's IOS XE base now. And we brought in all that 15 years of innovation, because this isn't a controller that you just put in one place. This is built for the most demanding networks that are out there to be really be the next generation of AirOs.
So this feature complete with new differentiated features from IOS XE, as well as new innovations like rolling AP upgrades. How do you update your access points without impacting the infrastructure or access from your users, because wireless just always needs to be on, I think that's the expectation for all users.
It also brings in the ability to run the wireless controller anywhere. So the 9800 can run as an appliance, which a lot of customers do that. But it can run virtually in a public cloud or on a virtual machine on-premise on a server of your choice. And then last but not least, it can actually be embedded in a switch.
So if you have a catalyst 9300 switch, it can now act as a wireless controller for up to 200 access points that are connected to it. So it's really a very flexible deploy anywhere, always on with the resiliency features and secure new wireless controller that we really think helps bring intent-based networking in a consistent way to wired and wireless.
Great, okay. I wanted to ask you about the new software features that have come on with the Cat9K. To some extent -- I mean, not to some extent to a larger extent this is a new concept for a lot of your customers out there that are used to using the catalyst line historically. And I'm just wondering, how the rate of adoption of the software features is going, and are there particular features or the switches you’d like to see being used more by the customers that are buying them?
I think we're clearly pleased with the traction on the Cat9K. We've talked about one of the fastest ramping product in the history of Cisco. And I think there is a few things that are really working well. One is, if you want to start with the new product and become intent-based networking ready, but deploy in the way that you're used to you can do that. So this is not a -- when you move the Cat9K, you have to do a completely new way of networking. Now it provide investment protection you can run the way you were running to go out.
Secondly, the OS consistently helps our customers certify and move more quickly. If you certify the 9300 buying a core switch with the 9500 or moving into wireless now becomes much more easy or much easier. The abstraction that we provide now where you are actually administrating network, managing the network through DNA Center are software layer, which is the controller layer. It actually helps you again, qualify one device and then roll out to very large network in a very consistent way.
So the whole idea of automation is to enable our customers to move more quickly and do it at lower cost and we're actually seeing this. But we're still early in the transition, but we're seeing this play out. Some of the most popular I would say, the services that our customers are rolling out as part of the architecture, number one is software-defined access. Almost every customer is looking to segment the network, whether it’s to address GDPR requirements, other compliance requirements, or is to stop the proliferation threats, like if your laptop is infected, make sure everybody inside your building or floor doesn't get infected it as well.
To enable researchers in a university to get access with their students to a new -- to a new application they are working on, but have nobody else access it. To have doctors in a hospital, get access to medical records, but nobody else can. To make sure the medical devices there can access their systems, but can't be compromised by hackers. Net certification and threat detection become really important, and that's where software-defined access is our architecture where simply with a few clicks you can turn on segmentation across a very, very large network in a fully automated way. So that's number one.
And number two is, something we're calling wireless assurance, which is the ability to route cause wireless issues using our partnership with Apple, so we have client data from the endpoint, we have data from the access points, the switches, the routers, from DACP, DNS other network services we talk to.
And we stitch it all together into a graph, into a contextual graph, where you can do a search and say hey, Sachin connecting right now or when he connected yesterday 5 o'clock, we maintain history, what's going on, what issues is he running into, like what happen why couldn't he connected the network or if he connected what was the issue, how do you go fix it.
This saves so much time for our customers, so much pain. And it helps them ensure that the wireless works in an awesome way. Works really, really well delivers a high quality experience. So these are two solution set we're really seeing a high-degree of interest in and we're going to continue to drive with our customers.
Okay, thanks. I guess, you would probably agree that the ramp of the Cat9K started a little lower than some other product trends, because I know you've lived through a few of these cycles now, having such a long career at Cisco. And I wonder, is -- do you think some of that is because of the software features at the learning curve of what is possible and then people have wanted to evaluate that before they purchase?
Actually, I don’t think we ever talked about Cat9K ramp starting to slower. I think, we've only talked about how it’s been the fastest ramping product in company history. And I think it goes back to the fact that the switch can run as a great switch the same way you were used to running a previous generation switch. And it also has all these new software capabilities.
And so, I think, that's remove a lot of the friction that you could maybe anticipate, where customers can buy this and say, look, I'm just going to buy the best new switch as bind to the architecture here. And they can look at rolling out the software capabilities the new architectural capabilities, when they are ready. So I don't think there's been -- with all the architectural innovation around it, I actually think that helps that ramp more than anything else.
Okay, that's fair. So you don't think it's really affected things because you can use it as a switch you would have had before or adopt the new software looking forward. What about education, I know Chuck has talked the need to educate users and I know that maybe not in this cycle, but as you look forward to the next time people replace switches, you're going to want to make sure that people are fully utilizing all these new software features. Or if you could talk a little bit Sachin about education and what -- maybe what you're doing differently in order to educate users as they buy the product or maybe before they buy the product or both?
Right, education I think is in -- we're looking at it in many angles, okay. One is first of all we obviously have to use all our traditional methods, like all the training, all the videos, all the cisco.com updates that we do. But in this case first of all we went all in with that kind of more traditional training through like I think we talked about at launch time tens of thousands of partners and Cisco salespeople who were trained on this new innovation. Because if you're going to reinvent networking you need to get the word out there, you need to get people feeling comfortable.
Now at the same time the new products are actually cloud tethered. So when you think about DNA Center, it connects with the cloud to get its updates to send telemetry data as well. That does a couple of things for us. One is, the information on the capabilities with the new releases the documentation for that can now also be achieved in products. That's an opportunity where we can educate through the product itself.
The second thing it does is, we can use the data we're getting to understand what customers are using and see what do we need to enhance. So DNA Center has a make a wish button on every single screen, where if a customer says, look I really love this wireless assurance capability, but I wish that this also showed me this additional data. They can make a wish, it comes straight to our engineering team and we can go enhance the product.
We put out new updates on DNA Center about every three to four weeks. This is the new world. It is a software world, delivered through the cloud, cloud updates. We did the same kind of thing on we manage which is in the cloud itself, right, for our SD-WAN Solution. And so the education here is a new SaaS type of model. It's not a educate once, you're done for a year or you're done for six months.
We're building capabilities here through communities, through video, through in product training, thinking like a SaaS company. And of course we have that products in Cisco, but this is about thinking in the core itself in enterprise networking like a SaaS company and driving education that way.
And if you -- that just out of curiosity have you allocated more of your budget to this sort of training in this product cycle or are you able to automate a lot of it and do it in a pretty low cost way?
I’m not sure, like about budget allocations, I think when we looked at -- I talked about we're going to increase the pace of innovation in the core and our pace of innovation across the board. I think we factored in the transformation that's needed everywhere in the company.
It's not just the products we're bringing out, but it is training, it is about as Partner Summit we talked about how our customer experience team CS team and our partner organization is helping move to a lifecycle approach. Really understanding where customers are in the adoption cycle helping them move from one step in that racetrack to another and move forward so that they can realize the full value of the software that they're buying.
And so this is -- I am not sure again not a budget answer, but certainly overall investment in the company moving in this direction we're all right.
Okay, all right. I want to change gears a little bit here and maybe talk about supply on the Cat9K, because we keep hearing from different VARS that at least historically looking back a quarter or two ago they couldn't get as much as they wanted. I wonder if you could just kind of talk us through maybe the history of supply of the product and where you stand now?
Look, so I think you mentioned earlier that I've seen many of these transitions, I mean, I've been on campus searching for over 15 years now and certainly you go through this kind of thing with any of your product ramp. We have a very, very sophisticated supply chain. And that supply chain is constantly managing the risks and all I can tell you is we have not seen any new surprises with the Cat9K. So for us this is business as usual and relying on that very sophisticated supply chain to do what's needed. And again nothing out of the ordinary.
So just to be clear, as we stand right now today, barring the launch of the 9200 and 9800 which I understand will be in volume supply by the end of December, I think, is what you guys have said, but you might want to correct that if that's not quite correct. But apart from that everything's in full supply now, there's no issue with any kind of supply shortage, everybody's able to get what they want. Is that true to say?
Yes, in general there we continue to be able to fulfill demand for the Cat9K and as you said for the 9200 and the 9800 we expect to ship those both in quarter.
Okay. And then could you just clarify to the 9200 and 9800 they've come later in the whole cycle the product cycle here, was that always part of your roadmap plan Sachin or I mean it's similar to what you've done in prior cycles? I'm not sure I remember or was that due to your anticipated demand for the products and wanting to make sure you had enough supply on hand for them?
No, I mean, let me explain it maybe it’s different answers for the 9200 and 9800. We've been working on these for a few years. Then the -- I mean, in general, the Cat3K class of the product has replaced for us. I mean, I think it's working, we've done that you come up with a high end with a capability and then you optimize and you bring out the one focus for mid-market and for branch. And I think we’ve follow some of our pattern there with the 9200 coming out after the 9300, if you will.
And so that was, I think, part of the roadmap planning that we had. But 9800 is a very different thing. Because here we had to spend a lot of energy to make sure that out of the gate this was a fully capable wireless controller. So this -- we're not running -- this is not an experimental component. This is a controller this breaking 15 years of innovation in AirOs. Again, remember AirOs has number one market share for wireless and on a phone, and now we're taking that operating system and we're going to be transitioning our customers to IOS XE where the Catalyst 9800. So that required a lot of care, a lot of feedback in the end from customers.
We've been there for running trials for several months. Again just to have it running with customers giving us feedback on do we have the right operational environment, does it integrate with DNA Center really well the way we expected so it’s fully automated to complete analytic on system. Does it have all the features they need?
So they can move from a controller they used for several years AirOs to the new one with confidence. So a little bit of a different story on between 9200 and 9800, but both have been the result and the fruits of our labor, if you will, after a few years.
Okay. Okay, great. That's helpful. I remember, but you probably don't remember this. But I think several years ago at a dinner one-time, you took me through the whole -- you gave me the one-on-one on Catalyst 9K refresh or catalyst refresh cycles and how long they last and how they ramp and all this sort of stuff. And I wonder, I kind of wanted to ask you whether you think this is a normal cycle.
I know that it took a little longer before you guys announced updates to the products that, I guess, that could affect maybe the ramp speed. But I'm just curious to know whether you think this will be sort of similar in terms of cycling length the prior catalyst to refresh cycles? Or do you think there's something different about this particular cycle?
I think, like historically you might have used speed as a reason and you're moving for faster Ethernet or get over the Ethernet and thought about that as a refresh cycle, right? And like -- honestly like, while there's an innovation cycle happening here, we don't really see it or think about it as a refresh. And so look, our customers are going out to upgrade, because they want to solve the problems. They want to solve certain problems for their business. And that gives us a chance to explain the value.
Look, if they're just out there to replace an old switch, because they've a really old switch they want a new switch, of course, the Cat9K is the best campus switch on the planet, they’re going to go do that. But that's not the maturity of our conversations. The maturity of our conversations now start with what -- how is access changing for you what's connecting?
Or what security and experience requirements does that present? How complex is it to deliver that security and experience? And now let's talk about intent based networking to give you the automation, the security and the analytics that you need to run this environment and run more quickly for your business.
And so I would say the number one conversation now is security. And the second conversation is probably about user and experience for all the things that are connecting, especially for wired and wireless. It's not a conversation that starts with refresh.
Okay. And so then it’d be fair to say I guess, you would expect the larger proportion of the install base to turnover or to upgrade to these new switches as a result of that because you are getting all the new features in software and so on. So it’s just not going to be the same as a normal refresh cycle maybe it takes longer or I mean, what do you think that means for, sort of, trough to peak on the product growth? Do you have -- and I think typically it's been about 18-month.
Sort of a perfect answer for you there, I would say two observations. I would say the things that I talked about in terms of the conversations we're having with customers are broad based. And I think the -- every customer is trying to solve an automation problem security problem things that I described. And our technologies and intent-based networking I think falls right at the heart of that. And so I think broad based conversations.
And the other thing I'll tell you is that -- or maybe two other things then that we've seen tremendous ramp in the products. So that gives us confidence that customers see the value and the value based purchase is working.
And now with the 9200 with the 9800, we've significantly expanded the type of customers that we can go after with intent-based networking. And so I think all of that put together says that look we think we're addressing what the market needs, what our customers need, but is it going to be fast or slower. I'm not sure I have data to give you a better answer.
Okay, that's great. Appreciate the attempt anyway. Let's switch gears to SD-WAN and talk a little bit about what you've been doing there. So you recently announced an integration of the SD-WAN and security portfolios and then an optimization for Microsoft Office 365. Could you explain why all that is important and how you feel like your SD-WAN is now differentiated from competition?
Yes, that's a long question so let me let me break that up a little bit. So first of all on SD-WAN, I think a few years ago, like, the buzz was hey, this helps you costs, it just helped you save on cost for WAN circuits. And I don't think that's the conversation anymore. Yes that's a part of it, but I think, the main conversation now is how you're consuming applications that's fundamentally changing.
So instead of you having your -- majority of your applications and data sitting in a private data center that you control, and all your remote branch sites can just connect over a secure network secure VPN, or a private sort of connection into that head end campus or data center. And then you ran your security stack writing from your data center. And life was pretty good.
Now what's happening is your users could be in the campus, could be the branch, could be in a coffee shop, could be anywhere. And they are wanting to access applications that could be in the private data center still, but a lot of those are now SaaS applications like Office 365, like, I mean, so many customers are looking at Officer 365 and other SaaS applications. Or they could be in a infrastructure as a service provider like AWS or Azure or Google Cloud.
And then our customers are now challenged with, hey how do I steer this traffic. Because even though the application sitting in the cloud my users are expecting a great experience. We're all -- we've all used to this as users and we don't care if we're using Salesforce or Box or O 365 doesn't matter, we expect an awesome experience. And our team needs to go figure out how to deliver the experience and how to deliver the security regardless of how the traffic flows.
So what we did is we took our SD-WAN solution with the big data and we took our best of breed security solution that we have in the security business group. So we took application firewall, we took URL filtering, we took ITS capability and we embedded that in our SD-WAN software offer. So now wherever you deployed, if you deployed virtually in the cloud or you deployed in the branch you can get the best of breed security stack automated with just a few clicks.
The second thing we did in security is we connected our cloud security with umbrella through to our SD-WAN solution. So if you need some of the traffic to go through umbrella in the cloud and then hit your cloud service itself, your cloud application itself, we can make that steering and the connection simple, one click in the Viptela if you manage you're ready to go. So that's on security side.
The other thing we did is we took API integration with Office 365 to help you pick the best path at that time to get to Office 365 to give you the best experience. And we all know from sort of modern mass applications that the shortest path may not be the best path. You may need to go around to actually get home quicker. And similarly, we're using intelligent data analytics and algorithms to figure out what is the best past, and that allows us to improve Office 365 experience way up to 40%.
So these things sort of working in concert, I mean, I think trying to tell the story. This is not about like just phasing WAN cost; this is about using all kinds of WAN circuits to achieve the best experience and again the best security for your users and for your applications. And I think you asked like how does this could differentiate? Well, what you need here is you need the best of breed routing stack, you still have to get traffic to the right place. You need to innovate with all kinds of environments, you need the best of breed security stack and you need the best of breed application experience stack. And we have all three.
And then the last thing you need by the way, because remember, that's where SD-WAN in the middle is just connecting users to applications, which means you need the campus network to be best-in-class. So you need all the innovations with intent-based networking in the campus that I described linked with the SD-WAN, linked with our data center with links like API. And so you need to solve a multi-domain problem. And I think this is a unique thing where only Cisco can solve this from a multi-domain and domain point of view.
Right, okay. On SD-WAN, can you talk little bit about the market opportunity? So should we be thinking about this as a similar market to your enterprise access routing market, or do you think it's bigger in some way. Could you just kind of help us understand how to think about sizing for the SD-WAN market opportunity?
Yes, this one is an interesting and I mean, there are analyst reports out there from Gartner and IDT for example that talk about this as a multi-billion dollar opportunity. I think the way we look at it is that this is an opportunity for software innovation that is delivered through the cloud is a SaaS model that we manage, that includes routing, security, application experience capabilities, where it's available in a physical form factor on our ISR, ASR portfolio, and in a virtual form factor using our ENCS which is basically like a server with a hypervisor in it or in the cloud itself completely virtualized.
And so, the way we think about it is the services are richer now that can be -- that there are software services that are deployed in many more places and not just the stack that runs in for data center, but in many, many more places that you need to steer traffic through. And so we do think it's an expansion of the overall opportunity, but how much it is a point is going up or down versus software going up or down like I don't think anybody is quite put their finger on. But overall, new problems that we are looking to solve with our customers that does present new opportunity.
Do you think -- I was just curious what you think about the future of this business. Do you think that the now that's speeds and feeds are getting less important that as you look forward, you could see this becoming almost a complete services kind of recurring revenue business overtime, which given what all of you're talking about? It almost sounds like that's the way it might be heading?
So all of our Viptela SD-WAN offers and in fact all of our Meraki SD-WAN offers are 100% subscription-based, like there is no other way to buy it. And similarly we fully automated solutions that I talked about with DNA Center on the campus side for wired, wireless with the Catalyst 9000 series for example. All of that is only available 100% subscription-based.
And so I think you're absolutely right, that this is a -- this is moving to a world of ongoing software innovation, where our customers are expecting this to be subscription-based, so there is an ongoing OpEx model. And it is -- that is exactly, where we are moving in our business. And you've already seen that with all the product introductions we’ve made in the enterprise networking business.
Yes. I was just thinking maybe ISR and ASR sales kind of don't -- people don't refresh them as much within that revenues flows toward the subscription services overtime more.
I think, I didn't comment on it earlier, but I'm not quite sure I agreed that speeds and feeds are no longer important. I mean everybody wants a faster circuit. If most of the applications are accessed through the internet you're going to want a richer more predictable internet connection, you may want two of them. And so your -- whatever appliance it is or server it is that you are putting out there needs to perform.
It also needs to perform not just for one thing like it may not just be doing routing, it may need to perform for security services. And so acceleration of those security services or having enough capacity to run those is going to be important. Same thing for application quality experience services that you might need to run.
So I think first of all I wouldn't say that speed and feeds are no longer report, I think that's not your primary driver. I think the software decision is your primary driver and once you're on a subscription and more of the value is in the subscription it could also be that replacing the underlying hardware is easier, right? Because let's say you bought a five year contract on a subscription and three years in you want to replace the router underneath because you upgraded your WAN circuit and you want something faster, while you can just bring in the new router and your subscription carry forward to the new router.
And so it's -- I think, it's a little bit to be determined on what it does from a hardware refresh point of view, but there's still reasons why -- again you don't want to refresh, but you want to upgrade the hardware because you're looking for additional capability.
Okay, yes it's interesting. What about access footprint. Can you say what proportion of the access footprint out there today uses SD-WAN?
I think look, both again from a DNA Center point of view and SD-WAN point of view, we're in very early days in the transition. I think we've talked about this in some of our calls as well that this is a multi-year transition that the market is going to go through that we need to help our customers go through. And so while we have a good number of customers adopting all of different solutions that we've put out there with DNA Center or with Viptela SD-WAN, we're in early days. So we don't share exact percentages, Rod, but there's a lot of runway here.
Okay. What about competition, Sachin, could you say a word about how the competitive environment looks? I mean, you guys acquired Viptela, VMware acquired VeloCloud. There's a bunch of independents out there. Talk to us a little bit about what you see out in the marketplace competition wise?
Yes, I think, some of the consolidation actually, it’s a little bit of a proof point to our strategy here and to the value that Cisco can bring and to the -- what our customers really trying to solve for, right. This is -- you can't just have capability in one area. It can't be that you have good enough routing or you have good enough security that's just not going to work. You need to have best of breed routing security and application quality experience capabilities.
You also need an awesome cloud delivered sophisticated automated solution for SD-WAN. So that's just in SD-WAN. And then you need to link all your domains wire to wireless access with SD-WAN, with the cloud, with your data center and not to mention you need security end-to-end. So all of our architectures can leverage, Talos for example. Our threat intelligence in the cloud so then we can refer to catch new threats and block those better than anybody else in the world.
And so when you think about that problem what is the customer after, they're after policy, what are my access rights, how we deliver security and experience like how do I ensure the mobile experience, the application experience is great, end-to-end. Regardless of where you connect from, regardless of where your application and data reside. That’s where the multi-domain solution here is important.
As I think some of the competitive landscape is going to try and evolve perhaps with some of this consolidation to achieve the outcome our customers are looking for. I think Cisco is uniquely positioned where we are the industry leaders in each of these domains and we can stitch these together and link them in a way that solves our customers’ problems in the best way.
Great. What about win rate, I mean, if we're talking about competition that another way to ask it. Like do you feel like your win rate is very high when you come up against competition or is it not really even a competitive situation a lot of times for you guys because you've already got the access router there.
So while we don't give the specific win rates. I mean we look at this obviously very closely. I can tell you that we are very happy with the execution of Viptela. We're very happy with the execution we have on the campus side for wired and wireless and I think we are showing up competitively differentiated and getting the preference from customers, both existing Cisco customers as well as potentially competitively how we count that we're transitioning.
As you know in routing, we also have very, very high share existing, but this is a problem that goes beyond routing and we think we're differentiated in the entire stack with SD-WAN as well as in our multi-domain approach.
Okay. And you mentioned the wireless, I'll just kind of lead on from that. Could you just describe the difference between the Viptela solution and the Meraki SD-WAN solution?
Yes, actually -- and look, both of those are needed to make sure we can address every part of the market. I mean, we're going for how we address the entire TAN [ph] here and so Meraki think of that as addressing the lean IT full stack management.
So I've got a very small team and I need to be able to manage switching, wireless, WAN, routing, security, surveillance cameras, maybe even MDM capabilities, all through a very small IT team. That's what the Meraki team offers. And their SD-WAN solution again has embedded security and it's managed through Meraki dashboard in the consistent way.
When you think Viptela I think, again enterprise flexible and sophisticated routing, advanced routing, secure segmentation maybe even more application experience more stringent needs there that's where you're coming to play, but you're likely to have a team that is more focused on just SD-WAN rather than an IT team that does everything.
And so, we've sort of all kinds of customers who have invested in IT differently and have different requirements for SD-WAN and I think together with the Cisco Viptela and the Cisco Meraki solutions we entered this whole market.
Do you think there's ever a day when those two solutions kind of merge under the same product umbrella? Or do you think it did make sense to just keep them separate indefinitely because of what you just described?
We do see so called hybrid deployment. So we do see some customers for example will buy a sort of enterprise network solution in the campus for example and maybe in the branch that actually went with Meraki wireless, right or they went with Meraki SD-WAN and DNA Center somewhere else in the campus. And so what we've actually done is we integrate from an operational point of view, where if you deploy Meraki somewhere the Meraki dashboard has open APIs that we interface through DNA Center so that you can get visibility to that infrastructure and start administering that infrastructure from DNA Center.
So think about DNA Center as a master in the case, where if a customer as a hybrid deployment, they can still operate and get visibility into their Meraki infrastructure from DNA Center. And I think that's the type of integration you can expect more off.
Okay. All right. That's fair. We now see enterprise access business has been a little bit slow since you guys acquired Viptela. Has that -- do you think the Viptela acquisitions caused people to pause at all as they evaluate the roadmap or is there anything -- is there any linkage between the SD-WAN and what's going on with enterprise routing?
Yes, I mean, I think what I'll tell you here is, we have plan when we acquired Viptela. Look, we're going to integrate the Viptela capability into our routers, it’s our routers, for example we have the plan to build up the security stack in it. And I think we're executing it very well against that plan. This is a significant change for customers, right? This is just like intent-based networking in the campus there's a new cloud edge, there's a massive transition here that one is happening and the world will take the time it needs to go and move forward in this direction.
We are investing again in training and pox helping our customers understand this and move forward, where we're pleased with the progress and I think our customers were excited that now they can roll up the solution on the routers that they already have because it works on our existing ISR, ASR family.
Okay, great. So we just published a big like 113 page report on 5G this morning. So I got to ask you a 5G question because of that. But I'm just curious, do you have this 5G change anything with your SD-WAN approach or does it -- just -- it's just another broadband mobile connection? Is there anything that you see shifting as 5G rolls out here in a couple of years?
So look, first of all I want to read your report, I'm going to ask Gloria for access to that, I'm curious. But I think from an SD-WAN point of view, we already see a lot of those routers have 4G LTE for backup or for as a primarily connection in some places as well. And so given 5G has faster speed, is expected to be a more reliable connection, we could see 5G being more as a backhaul, more of that uplink technology and of course as we put up new routers.
And when 5G is available, we would look to deliver that integration, just like we have 4G LTE solutions on both the enterprise routers as well as the Meraki MX family. And so, I think, that's the extent of the role that we will send right now that, but I’m going to read your report and learn more.
Okay, all right. I don't know how much you'll learn, but we'll definitely talk about it, if you want to. All right. So I wanted Kathryn if you could -- would you start polling the audience for questions just to see anybody has got any questions on the line and while you do that, I’ll ask one or two more to Sachin here.
Okay, so while we see if anybody queues up, Sachin, I wanted to just ask you about this whole subscription model. I mean, obviously your entire business -- you're the most affected by this, I think is Cisco right now. I'm just curious what sorts of responses or anecdotes you've seen with customers. I think anecdotes are always really interesting like discussions maybe you've had with actual customers who are considering this. So any thoughts in that respect any color that you can give us in terms of how that customer transition is going?
Yes, I mean, this is -- this one is like I think perhaps sometimes not as easy to understand on how we did this. And so a few things I like to share, first of all is. We did a lot of customer research before moving in this direction. We did content analysis, we looked at their willingness to pay for different capabilities, differentiation, automation, for example analytics and understood that there was a preference for a subscription OpEx model. If the innovation was delivered with net new, you can't just take the things that were perpetual before and moved into subscription. I mean that doesn't quite make sense.
If that was net new innovation that was coming out that perhaps for example helps them save on operational costs, which is an ongoing thing and a recurring thing or mitigate security threats which again could be an ongoing thing. Then there is a preference and a willingness to move to subscription.
And so we anchor our strategy based on that customer feedback and detailed research that we did and we wanted all our customers to try out the new intent-based networking capability or be entitled to go try it out. And so, we required three years of the base license essentials license as mandatory across the Catalyst 9000 series now for the wireless controller that's part of that family with the 9800 as well as our routers are using the Viptela SD-WAN solution.
And customers there have a choice, that, hey, if they believe in the value and have a plan to roll out they can buy them more the higher end packages. We've talked about how the majority have opted into buy the higher end package that tells us that they value the innovation we're bringing to market and if they're using and getting value they get to renew, otherwise they will have to renew. And so, I think it works well for customers which is pay for the value that you're getting, have an opportunity to buy up in the value if you want to use it. And we are saving you or providing value on an ongoing basis, which weren't a subscription.
Right, okay. There’s some along the line, there's this famous* military quote that “no battle plan survives first contact with the enemy.” I'm just curious like so you did all this market research, you had this plan, you went out there with subscription, it seems to be going well. What would you say the biggest surprises or change the thing that you didn't expect that that's come with the roll out of these subscriptions?
I think, look overwhelmingly I would say first of all be how smoothly we were able to drive the transition, I think was surprising. On the one hand first of all we have a huge business with Meraki that has been doing this for several years. And so the fact that we were able to take the enterprise networking business here perhaps shouldn't have been that bigger surprise. But again at your quote is spot on you want to make sure that you think through contingency plans and whatnot and we’ve looked at Meraki and we felt that we had enough data to move here and again so surprised, but has gone very, very smoothly.
Now once you go through it you get feedback on hey, how do I administer these licenses and I want to make sure that for as an enterprise I can see all my licenses in one place. I want to see what I'm entitled to, so management of the subscription needs to be simpler. So then we went in worked with our teams to focus on smart accounts, smart accounts and smart licensing is a capability where in cisco.com, you can log in as a customer and whenever you buy something from Cisco, it automatically gets into your smart account and you can look at all the licenses you're entitled to and manage them in a central place.
And so when you move to this new world, how you operate in a SaaS software world and manage licensing that was some of the learnings we had. So this is one example, it is like minor tweaks and other things that obviously we've had to do along the way, but broadly, what we sell out with we've managed to drive that pretty seamlessly.
Okay, great. Catherine, do we have any questions on the line?
You have one question coming from Kevin Stadler [ph]. Sir, your line is live.
Great. This is Kevin Stadler. So just if you think about the product roll out of the 9800, the AD version, which is geared towards a larger campuses, maybe describe that role at will you focus on that, will that be a kind of a try before you buy type of strategy? Or maybe just kind of, talk about the rollout of focusing on large campuses, environments, or would you go with introducing your, I guess, the 9800 40 product and see that the smaller markets, if you will.
So we actually think the -- when you're up at 80-gig or 40-gig, that's a pretty demanding environment. And yes, I mean, we're seeing customers look at those for maybe enterprise customers or universities that have whole bunch of wireless traffic that they have a centralized controller approach to, that's where the 9800 and AD S40 fit in, but then we at the same time introduced additional model modes to deploy. So you can take the 9800 software, the same software that runs on the 9800 AD and 40 appliances also runs virtually.
So we have a lot of customers saying, hey, I want to just try the new controller out. And they're able to get the software from Cisco downloaded into a virtual machine that they have, and now start seeing it and see the capabilities see how it performs. And for smaller rollouts, that may be all they do. Like they actually don't buy an appliance, they just run a server as a wireless controller.
And they also have customers who bought a 9300 switch, Catalyst 9300 or maybe looking to buy one, where they can run that same software. Again, as a containerized sort of virtual application that runs in the switch, like I have a small site where I bought a 9300 as an access point and the controller functionality for those access points can now run just on the 9300. So I don't think there's a one way to ramp this kind of thing happening, it's actually -- we have offers that service different site deployments that will ramp in parallel.
Just one other question, you talking about the subscription model what percentage of those contracts will be cancelled for convenience?
I don't know, I mean, I don't see that as something -- yes, I don't see that, let's put it that way. I don't see sort of a cancel for convenience approach.
So they are three-year licenses then. Okay.
Yes, three-year licenses are mandatory. Yes.
Okay. Great. Well, thank you very much.
Great. Thanks, Kevin. Catherine, do we have any other questions?
There are no further questions in queue at this time.
All right. I'm going to finish up with two then. Maybe, Sachin, say a word or two on how the channel partners have responded to subscriptions. You see a big difference between how some are embracing it and others are embracing it or do you see pretty ubiquitous kind of good reaction to it?
Yes, very quickly, I think we just came out of our Partner Summit had thousands of our partners there. Two things, one is the subscription model, I think everybody is fully on board on, so not a concerned there at all, in fact I think they’re embracing it fully. The second one is the opportunity that that creates for them from a lifecycle point of view. There, I think, people are or our partners are at different phases, some has built out full adoption lifecycle practices already, maybe they built it out with a different vendor, different partner or different part of Cisco and now see opportunity on the enterprise networking side.
And there is others who are much origin in the game. There's still mostly in acting as resellers or doing some racking and stacking and basic design. But aren't really looking at lifecycle adoption of intent-based networking and something they do best or now starting to look at that too. So those -- yes two part answer.
And then last question and then this can be a lightning round as John used to say. So what other products in your portfolio are left to transition to subscription? It seems like most of it's done, but what’s left to go?
I mean, I think, I can only speak to my business, which is the enterprise networking portfolio. You've seen SD-WAN move, you've seen identity services move, you've seen switching move, you've seen now parts of wireless move. And so from the enterprise networking business point of view, we're committed to moving to a consistent subscription-based offer across the portfolio.
Great. Okay. So you're pretty well done is the answer right on enterprise. You don't have anything else really material to do.
Well we -- I mean, we don't have all products we don't have our access points or routers win mandatory subscription, but we offer subscriptions across the board.
Right, okay. Great, well thanks, Sachin. Really appreciate your time. We're at the top of the hour so want to be respectful of your time, and as always, really interesting to talk to you. Appreciate you spending time with us today.
Thank you, Rod.
Thank you ladies and gentlemen, this does conclude today's conference call. You may disconnect your phone lines at this time and have a wonderful day. Thank you for your participation.