(energetic electronic music) >> Hello and welcome to a special CUBE presentation of the future of networking with Riverbed. I'm John Furrier, host of the CUBE. We're here with Paul O'Farrell, Senior Vice President and General Manager of SteelHead, Steelhead Connect. SD-WAN in action. Well, good to have you on The CUBE. Thanks for joining us. >> Great to be here. >> So, future of networking. This is something that we talk a lot about in our conversations, because the cloud's exploding, cloud business model. On-premise, true, private cloud. Hybrid, connecting to public clouds, is changing the game for app developers and large enterprises and how they do business. But it always comes back down to networking, 'cause everyone wants to know what's going on with networking. What is the future of networking? What's your perspective? >> Yeah, well John, as you said, everything's going to the cloud. But if you're a large multinational organization, you can't just click your finger and move your entire infrastructure to the cloud. But for the workloads that you do manage to move to AWS or Google Cloud or Azure, the good thing is that your IT organization is able to get out of the low-value-added activity of managing boxes and get into more strategic higher-impact activities and projects. So, if you think about moving a workload to the cloud, all of a sudden your organization is out of the business of managing boxes, managing servers, storage, and backup. But the challenge is that networking and the infrastructure required to connect all of that is still stuck in the past. And much of the way you manage a network really hasn't changed that much since the, certainly in enterprise networks, since the mid-90's when routers first really became popular. >> Give an example of why it's so hard, because I mean everyone wants networking to be faster. You have still move packets around the network. I mean boxes are changing. We know that the surveys are all pointing to non-differentiated labor being automated away. And that's clearly from the research. It's not a question of when, it's a question of when will, I mean not a question of how, when it's going to happen. So that puts pressure on the companies. When do they move from the manual networking to more automation? So give an example of some of the use cases. >> Yeah, so for a long time, as I said, the way you manage a network hasn't really changed. And in the last couple of years, we've seen the growth of a new market segment, or a new market, called software-defined WAN. So, taking some of the concepts of software-defined networking that had been trapped in the data center and then bringing those out onto the wide area network. And one of the big drivers was around the idea of, since there's so much more traffic going to the Internet, going to the cloud, I need a simpler way of managing that traffic. And I'd like to do it at a software level. I'd like to manage it based on policies and simple configurations that I could apply centrally as opposed to going down to the level of IP addresses and port numbers, as you have to in the sort of more traditional approach. So I think a lot of the initial impetus for people to look at new ways and new approaches to networking has been around this concept of direct-to-net, the desire to use more Internet transport, lower-cost Internet transport in the network. And that's sort of where it starts. And after that, you get to, what we at Riverbed believe is a bigger transformation of networking, which sort of begins with SD-WAN, but probably ultimately is more about really cloud networking. >> Some will say, and I'll get your reaction to this, that networking is outdated. Your thoughts? Is it outdated? Is it just moving too slow? Is it advanced? What are some of the, where's the progress bar on this conversation that's been kicking around the industry around networking needs to get updated and modernized? And is it outdated? What's your thoughts? >> Yeah, so as you said, at some level, you're always going to need networking, right? You've got to move packets around the network. You've got to connect applications with people and resources across the network. And it's particularly true in enterprises. But where I think the network has become stuck somewhat in terms of its evolution is that the traditional approach to configuring and managing devices, pre-staging routers and then shipping to a location where you have to do some more configuration on them, that piece of it I think has not evolved enough. But we're at a point now where a lot of the simplicity, the policy-based approach that you see in other parts of cloud infrastructure can now be applied to networking, that you can abstract away some of the complexity of the underlying network and then present that to an admin in a very simple fashion that looks very similar in terms of the experience to what happens when you deploy an application in the cloud, in AWS or Amazon. If you think about it, you can spin up an application and get it up and running in a matter of hours, if not minutes. You can deploy applications all over the world. Now, if you had asked somebody to do that 10 years ago, they would have looked at you like you're crazy. I want an application running in Frankfurt. And I want an application running in Seattle. And I want you to have it done by this afternoon, and by the way. Where it all falls down though is when I ask you to connect every root in my organization to those applications and have it done in a matter of hours or minutes. That's where it gets really hard using the traditional approaches. >> And, by the way, just to put in a point of clarification. I remember back when I was living in the 90's, 'cause what you described sounds like the 90's, that's a six-week project. Not like hours. That's like weeks. I got to make sure that the routers, we've got to configure the tables. All these manual efforts. But you're hitting on one of the things that is the future that people talk about, is really balancing the agility of doing something really fast, that's what the cloud is bringing to the table, with managing complexity. So that's one thread. So I want to talk about that. But also can I talk about the elephant in the room, which is, is my job going to go away? 'Cause, you know, a lot of those guys that are doing this command line interface stuff have built a job around their knowledge around configuring, which is not an agile. So they've got to be agile. So they're potentially at risk. So, future career. But the mandate of managing the complexity with agility. >> Yeah, so the industry obviously evolves over time. And, as you look at, again, go back to different parts of the infrastructure stack or the IT environment, you could have said just exactly the same, made exactly the same argument to me about servers and storage and backup administrators. Now, to my knowledge, those people haven't gone away. The total number of people working in the IT industry has not shrunk. If anything, it's grown significantly. So I think it's much more about freeing people from some of these laborious tasks that really don't add a lot of value and then redirecting those people to delivering on higher-impact initiatives. You know, a lot of talk in the industry these days, no matter actually what vertical you're in, about digital strategies, about transforming your business, and really what you want is to take your IT resources and your IT personnel and have them work on those projects and not have them-- >> John: The high-yield projects. >> Exactly. And to the extent possible, to automate a lot of the workflows and the way you manage day-to-day administration of the network, whether it's in the design phase, the deployment phase, or the management phase, of your network infrastructure, make that simpler and more intuitive and ultimately more like a consumer application, the types of workflows we're used to when we use web-based applications. Or perhaps, more reasonably, make it more like how you manage an application in AWS or Google Cloud or Azure. >> So your point about the server guys, the storage guys, their jobs never went away. First of all, there's more data coming than ever before, so they're always going to have a good job. So you're saying that is also applied to networking. >> Paul: Mm-hmm. >> It's still super important. >> Paul: Absolutely. >> And there's going to be more network, certainly with IOT on the horizon. You're going to have more connection points than ever before. So you're saying that tasks may go away, but the job will shift to other things, whether it's up the stack or other function that's related to adding value. >> Absolutely. So, the individual components that are deployed in the network that make the traffic, that allow the traffic to flow, that allow you to get the packets around the network, allow you to connect different parts of your enterprise, none of that goes away. But it just maybe takes a different form. And you mentioned IOT, for example. I mean that's a big question and a big challenge for a lot of organizations. How do you manage a network environment where you have more and more devices coming on the network? And instead of having, 10's, 100's of clients on a wireless network, for example, you could have 100's or 1,000's in a facility. And that's the type of new networking challenges that would be interesting to address as opposed to doing things that are, by their nature, manual and arguably can be done with a lot more automation. >> So I'm going to make a statement. And I want you to either agree or disagree or add some color to it. The future of networking is about automation, embracing automation to add value. And just as a point of validation, IOT, whatever trend that's happening right now that people get excited about, are all probably about machine learning. And everyone's saying that AI is going to solve the problem, which is simply just saying, technology's going to help with the automation. That's kind of my take on it. Your thoughts on that? Because that essentially is the validation. So the future of networking is, get used to automation. It's coming down the road pretty fast. >> So I think the first step towards taking some of that machine learning know-how and AI and applying it to networking is to automate networking. Make it easier. Make it policy-based. Don't make it about CLI commands. Make it about more manual configuration about scripting. The next step will be to apply machine learning and be able to have self-healing networks, being able to have networks which are aware of the types of-- >> Self-healing networks? Self-healing networks for having self-healing cars. Self-driving everything. I mean this is essentially the automation of what we're seeing. >> Sure, but let's start, let's not run before we can walk. Let's start with application-aware networks. How about that as an idea? Where at least the network doesn't think it's just passing packets, but actually knows what application it's using and is applying policies in an automatic fashion, whether it's to choose the optimal path for traffic or whether it's to apply security policies based on who the user is and what they're trying to do. So you should be able to do all that. And that is something that we built in our new product. >> Okay, so I would say that in hearing you, complexity is addressed by automation and software. >> Paul: Mm-hmm. >> The agility is really the application awareness of that. >> Yeah, I think that's a reasonable characterization of how to think about the future of networking, sure. >> Okay, so I want to get your thoughts on SD-WAN. We're hearing about that. With the cloud, and whether you're running true private cloud and hybrid and public, it's all an operating model. It's all a new way to think about provisioning networks and managing it. Isn't everything a WAN now? I mean, if you almost conceptually as a mind exercise say, the notion of local area networks and wide area networks are kind of, with the whole cloud thing, with the perimeter being decimated, and APIs flying around and microservices. I mean isn't everything a WAN now? >> Sure, I mean the whole concept of the WAN feels a little dated right now. I mean, if you think about it, if your kids are on the web or using their favorite social networking, and for some reason they can't get on the Internet, they rarely come down to you and say, "Hey Dad, the WAN's broken." So I mean clearly, people who live in the enterprise world still think in terms of wide area networks. But more and more, you're right. If you think about it, all of the different users who are coming on your network, whether employees or whether they're customers or partners, they're coming on using WiFi. There's a blurring of the line between the Internet, between the private enterprise network, traditionally referred to as the WAN, and the LAN. All of that is merging. And a lot of the technologies that Riverbed has been developing are really around this concept of SD-WAN, not just SD-WAN, but SD-LAN as well, and the ability to provide a single connectivity fabric across LAN, WAN, Cloud, and to the extent you still have data centers, most large enterprises will have those, data center as well. >> Great. And so competition. Let's talk about competition. You mentioned the CLI. Cisco's a market leader in all this. Your position vis-a-vis Cisco and how you look at the competition? >> Yeah, so Riverbed as a company has competed in various ways with large networking companies like Cisco for many, many years, since we started as a company. It is interesting that Cisco is trying to reposition itself, sees a need to change the way it delivers solutions for enterprise networking. It started by developing some of those capabilities within the organization and then more recently has made an acquisition of a startup, which we think is interesting, because it really validates the market now for SD-WAN. And we welcome it many ways, 'cause we think it's really the beginning of a shakeout and a maturing of the whole space. We think we're going to see that. >> You can't talk about the future of networking without talking about WiFi, because everyone who goes to a sporting event or concert, they lose their LTE, they go to the WiFi. And connectivity is like the lifeblood. You guys recently had an acquisition. What's the future like with you guys and WiFi? >> Yeah, we recently acquired a company called Xirrus, which was a company that had set out to build the fastest, most scalable, most, and this is really the key point, densest WiFi in the world in a very secure manner as well. And it was started by some of the pioneers of the WiFi industry, people who were in WiFI before it was even called WiFi. And so we thought they had an extraordinarily interesting technology. And what was particularly exciting about it is that they had also developed a cloud management approach to managing WiFi. As you said, WiFi is, on one hand you could think that WiFi is kind of a solved problem. It's been around for quite a while. But it's also become incredibly critical to not just enterprise networks, but to everyday life. We sometimes say that WiFi has become the inalienable right of every global citizen, good WiFi. If you think about the last time you checked in a hotel, what's probably the first thing you'd do is see if you can get on the WiFi. And if you have a bad WiFi experience, it doesn't matter how soft the Egyptian cotton on the sheets, on the bed is, >> John: It's plumbing. >> or how delicious the chocolate is on the pillow. >> People complain most about WiFi. >> Exactly. So if you think about it, some of our biggest customers now that we've entered the WiFi, and particularly the cloud-managed WiFi, business, our education, K through 12 and universities, on many university campuses, the users can have six, eight, 10 devices per person. Now, in the typical enterprise, maybe you don't have quite that many. But we're certainly all heading in that direction. And then you combine that with IOT and how people would like to put a lot of sensors and other devices on the network, then you're getting to a point where you really need incredibly dense WiFi. >> I mean IOT is about power and connectivity. WiFi gives great connectivity. Future of networking. Just summarize as we wrap this segment up for the folks watching who are practitioners in their jobs every day, trying to figure out the future, what's the bottom line of the future of networking? If you can give that statement and an example of how you guys are working with other practitioners. >> Sure. Well, first of all, I think the transformation that's occurring in the broader IT industry with the rise of the cloud and cloud networking and cloud computing is really extending now to the networking industry. A lot of the simplicity, the workflows, the automation and policy-based approaches is now extending to the space that network administrators have traditionally lived in. And I think that's really an opportunity for practitioners, as you call them, to really start using a set of more interesting and more capable tools that then will allow them to free themselves up from some of the lower-value-added activities to doing some really interesting things in the organization and to be an enabler of some of these new digital strategies and cloud strategies that their organizations are trying to execute. >> An example of companies you've worked with that might be a case study that you can share real quick? >> Well, what we're seeing is an awful lot of retailers, for example. So it's interesting. You see all the pressure that traditional retailers are experiencing from online e-commerce retailers. And what we're seeing is that more and more, they are using in-store WiFi. They're looking to put a lot more band-width into the stores to give customers in the store an incredibly compelling online experience while they're shopping. For two purposes. One is because they want to engage that customer while they're in the store. And also because they may want to do analytics and understand their behavior while they're in a store. But they want to do that at the same time as ensuring that some of their business-critical applications are up and running. So if you think about SD-WAN or cloud networking, it really provides the ability for us to do that, augment the WAN, deliver more band-width, lower cost band-width, into the store, but also give an incredibly compelling experience and have it all managed centrally with a simple policy-based approach. >> All right, the future of networking here at the CUBE Studio. Paul O'Farrell, Senior Vice President, General Manager at Riverbed. I'm John Furrier with the Cube. Thanks for watching.

(energetic electronic music) >> Hello and welcome to a special CUBE presentation of the future of networking with Riverbed. I'm John Furrier, host of the CUBE. We're here with Paul O'Farrell, Senior Vice President and General Manager of SteelHead, Steelhead Connect. SD-WAN in action. Well, good to have you on The CUBE. Thanks for joining us. >> Great to be here. >> So, future of networking. This is something that we talk a lot about in our conversations, because the cloud's exploding, cloud business model. On-premise, true, private cloud. Hybrid, connecting to public clouds, is changing the game for app developers and large enterprises and how they do business. But it always comes back down to networking, 'cause everyone wants to know what's going on with networking. What is the future of networking? What's your perspective? >> Yeah, well John, as you said, everything's going to the cloud. But if you're a large multinational organization, you can't just click your finger and move your entire infrastructure to the cloud. But for the workloads that you do manage to move to AWS or Google Cloud or Azure, the good thing is that your IT organization is able to get out of the low-value-added activity of managing boxes and get into more strategic higher-impact activities and projects. So, if you think about moving a workload to the cloud, all of a sudden your organization is out of the business of managing boxes, managing servers, storage, and backup. But the challenge is that networking and the infrastructure required to connect all of that is still stuck in the past. And much of the way you manage a network really hasn't changed that much since the, certainly in enterprise networks, since the mid-90's when routers first really became popular. >> Give an example of why it's so hard, because I mean everyone wants networking to be faster. You have still move packets around the network. I mean boxes are changing. We know that the surveys are all pointing to non-differentiated labor being automated away. And that's clearly from the research. It's not a question of when, it's a question of when will, I mean not a question of how, when it's going to happen. So that puts pressure on the companies. When do they move from the manual networking to more automation? So give an example of some of the use cases. >> Yeah, so for a long time, as I said, the way you manage a network hasn't really changed. And in the last couple of years, we've seen the growth of a new market segment, or a new market, called software-defined WAN. So, taking some of the concepts of software-defined networking that had been trapped in the data center and then bringing those out onto the wide area network. And one of the big drivers was around the idea of, since there's so much more traffic going to the Internet, going to the cloud, I need a simpler way of managing that traffic. And I'd like to do it at a software level. I'd like to manage it based on policies and simple configurations that I could apply centrally as opposed to going down to the level of IP addresses and port numbers, as you have to in the sort of more traditional approach. So I think a lot of the initial impetus for people to look at new ways and new approaches to networking has been around this concept of direct-to-net, the desire to use more Internet transport, lower-cost Internet transport in the network. And that's sort of where it starts. And after that, you get to, what we at Riverbed believe is a bigger transformation of networking, which sort of begins with SD-WAN, but probably ultimately is more about really cloud networking. >> Some will say, and I'll get your reaction to this, that networking is outdated. Your thoughts? Is it outdated? Is it just moving too slow? Is it advanced? What are some of the, where's the progress bar on this conversation that's been kicking around the industry around networking needs to get updated and modernized? And is it outdated? What's your thoughts? >> Yeah, so as you said, at some level, you're always going to need networking, right? You've got to move packets around the network. You've got to connect applications with people and resources across the network. And it's particularly true in enterprises. But where I think the network has become stuck somewhat in terms of its evolution is that the traditional approach to configuring and managing devices, pre-staging routers and then shipping to a location where you have to do some more configuration on them, that piece of it I think has not evolved enough. But we're at a point now where a lot of the simplicity, the policy-based approach that you see in other parts of cloud infrastructure can now be applied to networking, that you can abstract away some of the complexity of the underlying network and then present that to an admin in a very simple fashion that looks very similar in terms of the experience to what happens when you deploy an application in the cloud, in AWS or Amazon. If you think about it, you can spin up an application and get it up and running in a matter of hours, if not minutes. You can deploy applications all over the world. Now, if you had asked somebody to do that 10 years ago, they would have looked at you like you're crazy. I want an application running in Frankfurt. And I want an application running in Seattle. And I want you to have it done by this afternoon, and by the way. Where it all falls down though is when I ask you to connect every root in my organization to those applications and have it done in a matter of hours or minutes. That's where it gets really hard using the traditional approaches. And, by the way, just to put in a point of clarification. I remember back when I was living in the 90's, 'cause what you described sounds like the 90's, that's a six-week project. Not like hours. That's like weeks. I got to make sure that the routers, we've got to configure the tables. All these manual efforts. But you're hitting on one of the things that is the future that people talk about, is really balancing the agility of doing something really fast, that's what the cloud is bringing to the table, with managing complexity. So that's one thread. So I want to talk about that. But also can I talk about the elephant in the room, which is, is my job going to go away? 'Cause, you know, a lot of those guys that are doing this command line interface stuff have built a job around their knowledge around configuring, which is not an agile. So they've got to be agile. So they're potentially at risk. So, future career. But the mandate of managing the complexity with agility. >> Yeah, so the industry obviously evolves over time. And, as you look at, again, go back to different parts of the infrastructure stack or the IT environment, you could have said just exactly the same, made exactly the same argument to me about servers and storage and backup administrators. Now, to my knowledge, those people haven't gone away. The total number of people working in the IT industry has not shrunk. If anything, it's grown significantly. So I think it's much more about freeing people from some of these laborious tasks that really don't add a lot of value and then redirecting those people to delivering on higher-impact initiatives. You know, a lot of talk in the industry these days, no matter actually what vertical you're in, about digital strategies, about transforming your business, and really what you want is to take your IT resources and your IT personnel and have them work on those projects and not have them-- >> John: The high-yield projects. >> Exactly. And to the extent possible, to automate a lot of the workflows and the way you manage day-to-day administration of the network, whether it's in the design phase, the deployment phase, or the management phase, of your network infrastructure, make that simpler and more intuitive and ultimately more like a consumer application, the types of workflows we're used to when we use web-based applications. Or perhaps, more reasonably, make it more like how you manage an application in AWS or Google Cloud or Azure. >> So your point about the server guys, the storage guys, their jobs never went away. First of all, there's more data coming than ever before, so they're always going to have a good job. So you're saying that is also applied to networking. >> Paul: Mm-hmm. >> It's still super important. >> Paul: Absolutely. >> And there's going to be more network, certainly with IOT on the horizon. You're going to have more connection points than ever before. So you're saying that tasks may go away, but the job will shift to other things, whether it's up the stack or other function that's related to adding value. >> Absolutely. So, the individual components that are deployed in the network that make the traffic, that allow the traffic to flow, that allow you to get the packets around the network, allow you to connect different parts of your enterprise, none of that goes away. But it just maybe takes a different form. And you mentioned IOT, for example. I mean that's a big question and a big challenge for a lot of organizations. How do you manage a network environment where you have more and more devices coming on the network? And instead of having, 10's, 100's of clients on a wireless network, for example, you could have 100's or 1,000's in a facility. And that's the type of new networking challenges that would be interesting to address as opposed to doing things that are, by their nature, manual and arguably can be done with a lot more automation. >> So I'm going to make a statement. And I want you to either agree or disagree or add some color to it. The future of networking is about automation, embracing automation to add value. And just as a point of validation, IOT, whatever trend that's happening right now that people get excited about, are all probably about machine learning. And everyone's saying that AI is going to solve the problem, which is simply just saying, technology's going to help with the automation. That's kind of my take on it. Your thoughts on that? Because that essentially is the validation. So the future of networking is, get used to automation. It's coming down the road pretty fast. >> So I think the first step towards taking some of that machine learning know-how and AI and applying it to networking is to automate networking. Make it easier. Make it policy-based. Don't make it about CLI commands. Make it about more manual configuration about scripting. The next step will be to apply machine learning and be able to have self-healing networks, being able to have networks which are aware of the types of-- >> Self-healing networks? Self-healing networks for having self-healing cars. Self-driving everything. I mean this is essentially the automation of what we're seeing. >> Sure, but let's start, let's not run before we can walk. Let's start with application-aware networks. How about that as an idea? Where at least the network doesn't think it's just passing packets, but actually knows what application it's using and is applying policies in an automatic fashion, whether it's to choose the optimal path for traffic or whether it's to apply security policies based on who the user is and what they're trying to do. So you should be able to do all that. And that is something that we built in our new product. >> Okay, so I would say that in hearing you, complexity is addressed by automation and software. >> Paul: Mm-hmm. >> The agility is really the application awareness of that. >> Yeah, I think that's a reasonable characterization of how to think about the future of networking, sure. >> Okay, so I want to get your thoughts on SD-WAN. We're hearing about that. With the cloud, and whether you're running true private cloud and hybrid and public, it's all an operating model. It's all a new way to think about provisioning networks and managing it. Isn't everything a WAN now? I mean, if you almost conceptually as a mind exercise say, the notion of local area networks and wide area networks are kind of, with the whole cloud thing, with the perimeter being decimated, and APIs flying around and microservices. I mean isn't everything a WAN now? >> Sure, I mean the whole concept of the WAN feels a little dated right now. I mean, if you think about it, if your kids are on the web or using their favorite social networking, and for some reason they can't get on the Internet, they rarely come down to you and say, "Hey Dad, the WAN's broken." So I mean clearly, people who live in the enterprise world still think in terms of wide area networks. But more and more, you're right. If you think about it, all of the different users who are coming on your network, whether employees or whether they're customers or partners, they're coming on using WiFi. There's a blurring of the line between the Internet, between the private enterprise network, traditionally referred to as the WAN, and the LAN. All of that is merging. And a lot of the technologies that Riverbed has been developing are really around this concept of SD-WAN, not just SD-WAN, but SD-LAN as well, and the ability to provide a single connectivity fabric across LAN, WAN, Cloud, and to the extent you still have data centers, most large enterprises will have those, data center as well. >> Great. And so competition. Let's talk about competition. You mentioned the CLI. Cisco's a market leader in all this. Your position vis-a-vis Cisco and how you look at the competition? >> Yeah, so Riverbed as a company has competed in various ways with large networking companies like Cisco for many, many years, since we started as a company. It is interesting that Cisco is trying to reposition itself, sees a need to change the way it delivers solutions for enterprise networking. It started by developing some of those capabilities within the organization and then more recently has made an acquisition of a startup, which we think is interesting, because it really validates the market now for SD-WAN. And we welcome it many ways, 'cause we think it's really the beginning of a shakeout and a maturing of the whole space. We think we're going to see that. >> You can't talk about the future of networking without talking about WiFi, because everyone who goes to a sporting event or concert, they lose their LTE, they go to the WiFi. And connectivity is like the lifeblood. You guys recently had an acquisition. What's the future like with you guys and WiFi? >> Yeah, we recently acquired a company called Xirrus, which was a company that had set out to build the fastest, most scalable, most, and this is really the key point, densest WiFi in the world in a very secure manner as well. And it was started by some of the pioneers of the WiFi industry, people who were in WiFI before it was even called WiFi. And so we thought they had an extraordinarily interesting technology. And what was particularly exciting about it is that they had also developed a cloud management approach to managing WiFi. As you said, WiFi is, on one hand you could think that WiFi is kind of a solved problem. It's been around for quite a while. But it's also become incredibly critical to not just enterprise networks, but to everyday life. We sometimes say that WiFi has become the inalienable right of every global citizen, good WiFi. If you think about the last time you checked in a hotel, what's probably the first thing you'd do is see if you can get on the WiFi. And if you have a bad WiFi experience, it doesn't matter how soft the Egyptian cotton on the sheets, on the bed is, >> John: It's plumbing. >> or how delicious the chocolate is on the pillow. >> People complain most about WiFi. >> Exactly. So if you think about it, some of our biggest customers now that we've entered the WiFi, and particularly the cloud-managed WiFi, business, our education, K through 12 and universities, on many university campuses, the users can have six, eight, 10 devices per person. Now, in the typical enterprise, maybe you don't have quite that many. But we're certainly all heading in that direction. And then you combine that with IOT and how people would like to put a lot of sensors and other devices on the network, then you're getting to a point where you really need incredibly dense WiFi. >> I mean IOT is about power and connectivity. WiFi gives great connectivity. Future of networking. Just summarize as we wrap this segment up for the folks watching who are practitioners in their jobs every day, trying to figure out the future, what's the bottom line of the future of networking? If you can give that statement and an example of how you guys are working with other practitioners. >> Sure. Well, first of all, I think the transformation that's occurring in the broader IT industry with the rise of the cloud and cloud networking and cloud computing is really extending now to the networking industry. A lot of the simplicity, the workflows, the automation and policy-based approaches is now extending to the space that network administrators have traditionally lived in. And I think that's really an opportunity for practitioners, as you call them, to really start using a set of more interesting and more capable tools that then will allow them to free themselves up from some of the lower-value-added activities to doing some really interesting things in the organization and to be an enabler of some of these new digital strategies and cloud strategies that their organizations are trying to execute. >> An example of companies you've worked with that might be a case study that you can share real quick? >> Well, what we're seeing is an awful lot of retailers, for example. So it's interesting. You see all the pressure that traditional retailers are experiencing from online e-commerce retailers. And what we're seeing is that more and more, they are using in-store WiFi. They're looking to put a lot more band-width into the stores to give customers in the store an incredibly compelling online experience while they're shopping. For two purposes. One is because they want to engage that customer while they're in the store. And also because they may want to do analytics and understand their behavior while they're in a store. But they want to do that at the same time as ensuring that some of their business-critical applications are up and running. So if you think about SD-WAN or cloud networking, it really provides the ability for us to do that, augment the WAN, deliver more band-width, lower cost band-width, into the store, but also give an incredibly compelling experience and have it all managed centrally with a simple policy-based approach. >> All right, the future of networking here at the CUBE Studio. Paul O'Farrell, Senior Vice President, General Manager at Riverbed. I'm John Furrier with the Cube. Thanks for watching.