(upbeat music) >> Hey, welcome back to theCUBE's cover of DockerCon Mainstage, I'm John Furrier, host of theCUBE. We're here with Shubha Rao, Senior Manager, Product Manager at AWS, in the container services. Shubha, thanks for coming on theCUBE. >> Hi, thank you very much for having me, excited to be here. >> So obviously, we're doing a lot of coverage with AWS recently, on containers, cloud native, microservices and we see you guys always at the events. But tell me about what your role is in the organization? >> Yeah, so I lead the product management and developer advocacy team, in the AWS Container Services group, where we focus on elastic containers. And what I mean by elastic containers, is that, all the AWS opinionated, out of the box solutions that we have for you, like, you know, ECS and App Runner and Elastic BeanStalk. So where we bring in our services in a way that integrates with the AWS ecosystem. And, you know, my team manages the product management and speaking to customers and developers like you all, to understand how we can improve our services for you to use it more seamlessly. >> So, I mean, I know AWS has a lot of services tha t have containers involved with them and it's a lot of integration within the cloud. Amazon's as cloud native as you're going to get at AWS. If I was a new customer, where do I start with containers if you had to give me advice? And then, where I have a nice roadmap to grow within AWS. >> Yeah, no, that's a great question a lot of customers ask us this. We recommend that the customers choose whatever is the best fit for their application needs and for their operational flexibilities. So, if you have an application which you can use, pretty abstract and like end to end managed by AWS service, we recommend that you start at the highest level of abstraction that's okay to use for your application. And that means something like App Runner, where you can bring in a web application and run it like end to end. And if there are things that you want to control and tweak, then you know, we have services like ECS, where you get control and you get flexibility to tweak it to your needs. Be it needs of like, integrations or running your own agents and running your own partner solutions or even customizing how it scales and all the, you know, characteristics related to it. And of course we have, if there are a lot of our customers also run kubernetes, so that is a requirement for you, if your apps are already packaged to run, you know, easily with the kubernetes ecosystem, then we have, yes, for you. So, like application needs, the operational, how much of the operations do you want us to handle? Or how much of it do you want to actually have control over. And with all that, like the highest level of abstraction so that we can do the work on your behalf, which is the goal of AWS. >> Yeah, well, we always hear that all that heavy lifting, undifferentiated heavy lifting, you guys handle all that. Since you're in product management, I have to ask the question 'cause you guys have a little bit longer view, as you have think about what's on the roadmap. What type of customer trends are you seeing in container services? >> We see a lot of trends about customers who want to have the plugability for their, you know, services of choice. And our EKS offerings actually help in that. And we see customers who want an opinionated, you know, give me an out of the box solution, rather than building blocks. And ECS brings you that experience. The new strengths that we are seeing is that a lot of our customer workloads are also on their data centers and in their on-prem like environments. Be it branch offices or data centers or like, you know, other areas. And so we've recently launched the, anywhere offerings for you. So, ECS anywhere, brings you an experience for letting your workloads run and management that you control, where we manage the scaling and orchestration and the whole like, you know, monitoring and troubleshooting aspects of it. Which is the new trend, which seems to be something that our customers use as a way to migrate their applications to the cloud in the long term or just to get, you know, the same experience and the same, like, constructs that they're familiar with, come onto their data centers and their environments. >> You know, Shubha, we hear a lot about containers. It's becoming standard in the enterprise now, mainstream. But customers, when we talk to them, they kind of have this evolution, they start with containers and they realize how great it is and they become container full, right. And then you start to see kind of, them trying to evolve to the next level. And then you start to see EKS come into the equation. We see that in cloud native. Is EKS a container? Or is it a service? How does that work with everything? >> So EKS is a Amazon managed service, container service, where we do the operational set up, you know, upgrades and other things for the customer on their behalf. So basically, you get the same communities APIs that you get to use for your application but we handle a little bit of the integrations and the operations selected to keeping it up and running with high availability. in a way that actually meets your needs for the applications. >> And more and more people are dipping their toe in the water, as we say, with containers. What are some of the things you've seen customers do when they jump in and start implementing that kind of phase one containers? Also, there's a lot of head room beyond that, as you mentioned. What's the first couple steps that they take? They jump in,, is it a learning process? Is it serverless? Where is the connection points all come together? >> Great, so, I want to say that, no one solution that we have, fits all needs. Like, it's not the best case, best thing for all your use cases, and not for all of your applications. So, how it all comes together is that, AWS gives you a ecosystem of tools and capabilities. Some customers want to really build the, you know, castle themselves with each of the Lego block and some customers want it to be a ready made thing. And I want, you know, one of the things that I speak to customers about is, is to rethink which of the knobs and controls do they really need to have, you know because none of the services we have is a one way door. Like, there is always flexibility and, you know ability to move from one service to the other. So, my recommendation is to always like, start with things where Amazon handles many of the heavy lifting, you know, operations for you. And that means starting with something like, serverless offerings, where, like, for example, with Lambda and Forget, we manage the host, we manage the patching, we manage the monitoring. And that would be a great place for you to use ECS offering and, you know, basically get an end to end experience in a couple of days. And over time, if you have more needs, if you have more control, you know, if you want to bring in your own agents and whatever else you have, the option to use your own EC2 Instances or to take it to other, like, you know, parts of the AWS ecosystem, where you want to, you know, tweak it to your needs. >> Well, we're seeing a lot of great traction here at DockerCon. And all the momentum around containers. And then you're starting to get into trust and security supply chain, as open source becomes more exponentially in growth, it's growing like crazy, which is a great thing. So what can we expect to see from your team in the coming months, as this rolls forward? It's not going away anytime soon. It's going to be integrated and keep on scaling. What do we expect from the team in the next month or so? Couple of months. >> Security and, you know, is our number one job. So you will continue to see more and more features, capabilities and integrations, to ensure that your workloads are secure. Availability and scaling are the things that we do, you know, as keep the lights on. So, you should expect to see all of our services growing to make it like, more user friendly, easier, you know, simpler ways to get the whole availability and scaling to your needs, better. And then like, you know, very specifically, I want to touch on a few services. So App Runner, today we have support for public facing web services. You can expect that the number of use cases that you can meet with app runner is going to increase over time. You want to invest into making it AWS end to end workflow experience for our customers because, that's the easiest journey to the cloud. And we don't want you to actually wait for months and years to actually leverage the benefits of what AWS provides. ECS, we've already launched our, like, you know, Forget and Anywhere, to bring you more flexibility in terms of easier networking capabilities, more granular controls in deployment and more controls to actually help you plug in your preferred, you know, solution ties. And in EKS, we are going to continue to keep the communities, you know, versions and, you know, bring simpler experiences for you. >> A lot of nice growth there, containers, EKS, a lot more goodness in the cloud, obviously. We have 30 seconds left. Tell us what you're most excited about personally. And what should the developers pay attention to in this conference around containers and AWS? >> I would say that AWS has a lot of offerings but, you know, speak to us, like, come to us with your questions or, you know, anything that you have, like in terms of feature requests. We are very, very eager and happy to speak to you all. You know, you can engage with us on the container store map, which is on GitHub. Or you can find, you know, many of us in events like this, AWS Summits and, you know, DockerCon and many of the other meetups. Or find us on LinkedIn, we're always happy to chat. >> Yeah, always open, open source. Open source meets cloud scale, meets commercialization. All happening, all great stuff. Shubha, thank you for coming on theCUBE. Thanks for sharing. We'll send it back now to the DockerCon Mainstage. I'm John Furrier with theCUBE. Thanks for watching. (upbeat music)

>> From San Francisco, it's theCube, covering Girls in Tech Catalyst Conference, brought to you by Girls in Tech.. >> Hey, welcome back everybody. Jeff Rick here with theCUBE. We're in downtown San Francisco at the Girls in Tech Catalyst Conference 2018. About 700 mainly women, a few men, and they just brought in a busload of kids to come in and hear inspirational stories, really of women in technical leadership positions for the last two days. It's a really great event. We're excited to be back for a second time, and our next guest who's been travelin' just as much as theCube team, all the way back from Cisco Live, which we were at yesterday. Shubha Govil, she's the director of product management from Cisco. Welcome. >> Thank you. >> So how was Cisco Live? >> Cisco Live was awesome. Yes, we had very different audience there. Percentage of women, as you can imagine, sometimes in the networking industry, is not as much as we would like to see, but that's why we are here. >> That's why you're here. >> And we are really trying to bring in lot more women in our product teams within Cisco. Again, it comes down to having the 50/50 voice, so this is a great place to be to meet other like-minded women, in industry, and trying to get some talent. >> Well, good for you for making the trip, 'cause I'm sure you guys are wiped out. When it's your own show you work harder than any other show, but it really begs a question. How long has Cisco been involved in Girls in Tech, and, again, what is it that this gives you that's so different than a big conference? >> Correct, so Cisco has been involved for last few years for Girls in Tech, and just like we have been involved with several other organizations in the industry, really it comes down to being out there and spotting the talent. Big part of that is being at the events, and networking with the talent, and understanding their needs. This comes down to really finding the right perspective, as well as the cultural fit for people that we bring in. The best part about the Girls in Tech events is that they're a lot more hands-on training that they are doing, in terms of as part of Cisco's DevNet environment. So I'm part of Cisco's DevNet team, and we are driving Cisco's developer program, to build more on top of Cisco's APIs, and in that role, always looking for people who are ready to go hands on, and build cool solutions on top of Cisco API, so this is a great place to have been doing a lot of coding camps, and other formal boot camps where girls can come in, and be part of this ecosystem, be ready for the next opportunity that comes. >> It's interesting because you can't just do what you could do in the past, which is just go do the campus recruiting, and kind of the things that we think of as everyday HR pipeline, because you need more, and you need more diverse. So to be active in all these various organizations that have very strong focuses in diversity, whether it be women or unrepresented populations, et cetera. So, pretty interesting investment that you guys are making there, with time, money, and people. >> Absolutely, absolutely. It is key. It is totally the key for Cisco, and for every single technology company out there to be out there and finding the diverse opinion. It really comes down to technologies not made just for men, and 50% of female population, there's not a lot of thinking that goes in in designing the technology as to how different people will use it, and big part of it is bringing the people who can think from that perspective, and that's kind of where we are out there, making sure that we can bring in that opinion. >> And the culture's such a bit thing, and you guys had such a big culture change with the new CEO shift, because there's such strong personalities, and now you guys have moved onto a new CEO. We keep hearing about culture over and over again, and how important it is to bring that up. So how important is it from the cultural aspect to be involved, and get these diversity of opinion? >> It is huge. So really, Chuck Robbins has been bringing a very humble culture, people really trying to be there for everyone, each other. And as a committee, you are really building the talent, not just for doing the right jobs, and bringing the right perspective, but also culturally bringing those opinions, as well as bringing the thinking that's going to change the culture moving forward. The technology disruption that's around us has to do a lot with how, culturally, things are changing. There is amalgamation of people coming from all over the world, and in that mode, when you're designing a technology, or when you are from networking perspective, as we think about Cisco's networking culture, network engineers are evolving too, and they are becoming more part of programmable network, and that culture shift goes along with it, which is to bring in the right people culture, and part of it is being out at the events, and meeting people coming from different places, and bringing those opinions. >> It's interesting we're at Bespoke, because I think it was last year we were here with the Cisco DevNet Team, and really a different kind of point of view coming out of Cisco, led by Susie Wee and the team, in terms of reaching out for developers, not a closed system, really trying to engage with the developer community. >> And that's part of it. Cisco's DevNet Committee is, we recently crossed a milestone of 500,000 developers. >> Oh, so I heard that was the big celebration at the party, right? Half a million, very good. >> Big celebration at Cisco Live. >> Congratulations. >> And DevNet Create was an attempt in that direction as well, to really bring the application developer, and that thinking about network engineers who have been changing the way each application works, how the internet of things is going to further drive the growth of internet, in that world, we also need a lot of application developers coming in, and that was the attempt for DevNet Create Conference, and that's where Girls in Tech and other such events are very important. >> Right, and only going to be more crazy when 5G comes online in a couple years. The demand for networking is, and the bandwidth is not slowing down anytime soon. >> It's not. (laughs) >> All right, Shubha, well thank you for spending a few times, I'm sure you are tired after the long event, so hopefully you get through this and you can take it easy this weekend. >> Thanks, Jeff. It was good to talk to you. >> All right, thanks for stopping by. I'm Jeff, and you're watching theCUBE from Girls in Tech Conference 2018. Thanks for watching. (laid-back electronic music)

>> From San Francisco, it's theCube, covering Girls in Tech Catalyst Conference, brought to you by Girls in Tech.. >> Hey, welcome back everybody. Jeff Rick here with theCUBE. We're in downtown San Francisco at the Girls in Tech Catalyst Conference 2018. About 700 mainly women, a few men, and they just brought in a busload of kids to come in and hear inspirational stories, really of women in technical leadership positions for the last two days. It's a really great event. We're excited to be back for a second time, and our next guest who's been travelin' just as much as theCube team, all the way back from Cisco Live, which we were at yesterday. Shubha Govil, she's the director of product management from Cisco. Welcome. >> Thank you. >> So how was Cisco Live? >> Cisco Live was awesome. Yes, we had very different audience there. Percentage of women, as you can imagine, sometimes in the networking industry, is not as much as we would like to see, but that's why we are here. >> That's why you're here. >> And we are really trying to bring in lot more women in our product teams within Cisco. Again, it comes down to having the 50/50 voice, so this is a great place to be to meet other like-minded women, in industry, and trying to get some talent. >> Well, good for you for making the trip, 'cause I'm sure you guys are wiped out. When it's your own show you work harder than any other show, but it really begs a question. How long has Cisco been involved in Girls in Tech, and, again, what is it that this gives you that's so different than a big conference? >> Correct, so Cisco has been involved for last few years for Girls in Tech, and just like we have been involved with several other organizations in the industry, really it comes down to being out there and spotting the talent. Big part of that is being at the events, and networking with the talent, and understanding their needs. This comes down to really finding the right perspective, as well as the cultural fit for people that we bring in. The best part about the Girls in Tech events is that they're a lot more hands-on training that they are doing, in terms of as part of Cisco's DevNet environment. So I'm part of Cisco's DevNet team, and we are driving Cisco's developer program, to build more on top of Cisco's APIs, and in that role, always looking for people who are ready to go hands on, and build cool solutions on top of Cisco API, so this is a great place to have been doing a lot of coding camps, and other formal boot camps where girls can come in, and be part of this ecosystem, be ready for the next opportunity that comes. >> It's interesting because you can't just do what you could do in the past, which is just go do the campus recruiting, and kind of the things that we think of as everyday HR pipeline, because you need more, and you need more diverse. So to be active in all these various organizations that have very strong focuses in diversity, whether it be women or unrepresented populations, et cetera. So, pretty interesting investment that you guys are making there, with time, money, and people. >> Absolutely, absolutely. It is key. It is totally the key for Cisco, and for every single technology company out there to be out there and finding the diverse opinion. It really comes down to technologies not made just for men, and 50% of female population, there's not a lot of thinking that goes in in designing the technology as to how different people will use it, and big part of it is bringing the people who can think from that perspective, and that's kind of where we are out there, making sure that we can bring in that opinion. >> And the culture's such a bit thing, and you guys had such a big culture change with the new CEO shift, because there's such strong personalities, and now you guys have moved onto a new CEO. We keep hearing about culture over and over again, and how important it is to bring that up. So how important is it from the cultural aspect to be involved, and get these diversity of opinion? >> It is huge. So really, Chuck Robbins has been bringing a very humble culture, people really trying to be there for everyone, each other. And as a committee, you are really building the talent, not just for doing the right jobs, and bringing the right perspective, but also culturally bringing those opinions, as well as bringing the thinking that's going to change the culture moving forward. The technology disruption that's around us has to do a lot with how, culturally, things are changing. There is amalgamation of people coming from all over the world, and in that mode, when you're designing a technology, or when you are from networking perspective, as we think about Cisco's networking culture, network engineers are evolving too, and they are becoming more part of programmable network, and that culture shift goes along with it, which is to bring in the right people culture, and part of it is being out at the events, and meeting people coming from different places, and bringing those opinions. >> It's interesting we're at Bespoke, because I think it was last year we were here with the Cisco DevNet Team, and really a different kind of point of view coming out of Cisco, led by Susie Wee and the team, in terms of reaching out for developers, not a closed system, really trying to engage with the developer community. >> And that's part of it. Cisco's DevNet Committee is, we recently crossed a milestone of 500,000 developers. >> Oh, so I heard that was the big celebration at the party, right? Half a million, very good. >> Big celebration at Cisco Live. >> Congratulations. >> And DevNet Create was an attempt in that direction as well, to really bring the application developer, and that thinking about network engineers who have been changing the way each application works, how the internet of things is going to further drive the growth of internet, in that world, we also need a lot of application developers coming in, and that was the attempt for DevNet Create Conference, and that's where Girls in Tech and other such events are very important. >> Right, and only going to be more crazy when 5G comes online in a couple years. The demand for networking is, and the bandwidth is not slowing down anytime soon. >> It's not. (laughs) >> All right, Shubha, well thank you for spending a few times, I'm sure you are tired after the long event, so hopefully you get through this and you can take it easy this weekend. >> Thanks, Jeff. It was good to talk to you. >> All right, thanks for stopping by. I'm Jeff, and you're watching theCUBE from Girls in Tech Conference 2018. Thanks for watching. (laid-back electronic music)

(upbeat music) >> Live from Orlando Florida. It's the Cube. Covering Cisco Live 2018. Brought to Cisco NetApp and the Cube's ecosystem partners. >> Welcome back to the Cube's Live coverage here in Orlando Florida for Cisco Live 2018. It's the Cube's coverage. >> I'm John Furrier. The host. Here for three days of wall-to wall-coverage. Our next guest is Shubha Govil. Whose the director of product management for Cisco DevNet. Welcome to the Cube. Thanks for coming on. >> Thanks John. Thanks for having me on Cube. >> Great conversation before the cameras came out. We're talking about development and Cloud Native. But we're super impressed with the work you guys have done at DevNet. Certainly it's the top story of the show here is that Cisco has now crossed over the flywheel of innovation where 500,000 registered developers. >> Developers. Not visitors to a website. >> Yes. Or some marketing program. >> Correct. >> Real engaging developers. >> Absolutely. >> Congratulations. >> Thank you John. Thank you. >> A couple of years. Four years and you're here. You've got DevNet. And DevNet Create. Which we've been covering extensively as well. >> Yes. >> Which is the Cloud Native world coming together. >> Yes. >> This is for the first time in Cisco's history where you have now a clear line of sight for network engineers network developers network experts who have been certified in the CCIE and other certifications. CCNA, CCNP. All the stuff you guys do. You can now see a clear line where you can extend the capabilities and knowledge and expertise in power of networking. >> Absolutely. >> Up the stack. >> Absolutely. >> Finally Cisco's moving up the stack. >> Yes. >> Tell us what's going on in the product side? >> Yeah Absolutely. And I'm gonna talk about very specific example today. And today if you heard the keynote speak. And Susie opened up a few things and announcement. One of them was DNA developer center. And I'm gonna talk about that. Because part of it is how network is an open platform now. And that was part of the announcement. You will hear a lot about that. And linking it back to Dev reports. Quite right. They should care about why they should look into it. So three things I'm gonna talk about. DNA, developer center and what they can find there. And once they go there and they really start learning about our platform on API's on DevNet. What cord exchange does for them. And how they can start not only programming the Intent based on our Intent API's and what they want network to do for them but also sharing some community cord. Are using that community cord. Community Cord if they are just getting started. Right? So on DNA Developers Center we have four capabilities highlighted. These are the API's. Whether they are Intent API's Integration API's to connect with other third parties. Or SDK's to manage multi-party devices. Or there for ITSM or a specific use case integrations. >> So hold up. Go slow. >> I'm kind of not on the uptake as you are on this. Because you're in it. IF DNA is a set of abstractions API's on top of the equipment. >> Correct. >> So it's not natively. It's a set of API. >> Set of API's. >> So that people could use those API's to create services. >> To manage their network wealth. To automate and drive these right use cases. So I might. >> Give me an example. >> Yeah let's talk about an example. Intent. My intent might be to. We were talking about radio conferences awhile back. And I come from that environment. I want to drive a Butler QS for certain level of execs. Right? If they are on the call this was the thing of like eight ten years back. If my execs are on a call make sure they have the best experience. So the QS quality of your network should be set up to a level that there's no disruption. There's no latency in the call. Right? So that's an intent. That's a business intent. Give best experience to my execs. >> So really that's combining policy and QOS together to make it meet the outcome. Which is no latency. >> To meet the outcome. But for the network engineer now let's connect back to the developer. The network engineer whose trying to make this intent possible for the execs. There's a places they need to set up the SQSS. And won't it be easy of them if there was a simple API that they can use to create that solution to drive that policy across the devices. Whether Cisco devices or non Cisco devices. >> This has been the challenge for network engineers in general. Because you want to have things in control and locked down but as you want to do more things that are programmable. >> Correct. >> There's been some provisioning and some configuration management things. >> Correct. >> You're saying, hey you're gonna lock down all the architecture and then move up. Use the API's to do better integration. Make things run smoothly without disrupting the network. Is that right? >> That's part of it. But also it's about making it easy for them. Correct? Simplify the process of doing it. The process of making it happen was long steps of CLI command. That now that network engineer was going continuously. A lot of the time people actually tells us that they would have this cut and paste copy of the command. That they will take from one place go to the next place next device and next device. And continue to do that step. And that's the productivity game we are driving by simplifying where one API call can go across all the devices and make that change happen. >> We've heard that a lot from on DevNet and the hallway conversations that said DevNet's made my life easier. >> Yes. >> I don't have to do those mundane tasks. >> Exactly. >> That were part of getting things done. Okay. Let me ask you personal question. As director of product management for DevNet. What is your product scope? What are you working on? Can you take a minute? >> That's a very good question. And that's where some of these offers we were talking about earlier come in to play. So for example, within Devnet we create a lot of offers to make developers lives simple. Whether we are talking about giving them the best quality of learning content. Or giving them hosted Sandbox environment to try and test. All of that requires a lot of product management knowledge and the need. But really what the 2ADS we have work more closely to get them out to market. One is the thing called Code Exchange. It's a tool for our developer committee. Where we have aggregated the public git code across the Cisco technologies. >> That's on GitHub I think. >> GitHub code right. Absolutely. But the second powerful thing on top of that is our Ecosystem Exchange. This is where we are bringing an aggregated view of every partner out there. Every Cisco partner whose creating great solutions on our API's in a single place our developers can go and find that solution. To really address the business outcome they are looking to address. >> Shubha, I want you to put some color commentary around of some of the feedback you've heard. We hear people of the DevNet community saying I've come to Cisco Live and I spend all my week here in DevNet. Because it really is kind of like a kid in the candy store. (Shubha laughs) >> From a computer science or developer prospective. >> Yeah. >> What are some of the cool examples and demos that you guys have here? What's your favorite? What are some of the things that are jumping out that people are gravitating towards? >> I will tell you one of the most popular sessions that I have seen in the last few days here is Network Programmability for Networking Juniors. That's one. There's also a very Network Programmability one-on-one. Coding one-on-one class. It's basic Python. But applying it in network context. Those are some of the most popular sessions that I have seen. But when it comes to cool demos there's a cool demo around Flex IQ. I think you might be talking to Ashish later about that. >> Yes. >> And really it's a retail scenario how you are tracking. Using the location based service example. But in this case camera feed. Really analyzing where people are. And you'll get to hear more about this. >> We took a ad. I saw the demo. >> Yeah. >> The Flex IQ. First of all I love the name. I said trademark it immediately. (Shubha laughs) Get it out there. First use wins. And it's already out there. But it's really taking a A access point. >> This is an access point. >> And it plugs into a camera. And a great example of some of the coolness you can do with a preexisting condition. In this case an access point. >> So each of these information points that data one that they are collecting. Whether it's a camera feed. It's a location service. Like information about the devices and the environment. Each set of data is the relevance in this. Which is driving the newest use cases. And this data is coming through API's that have labeled but I'd say morockie access point API. All the camera API that are labeled that have enabled C Space. >> This is really the aha moment for me. I've been following Cisco really since the 90's. >> Yeah. >> Or at least when they formed. Being the young gun at the time. Younger than I am now. 30 years ago. But it was really networking. Connecting companies together. It was the plumbing. It was the core. >> Yes. Unstoppable since then. Now the success is still there. But it's really the problem solving is never going away. I saw this security challenges that were outlined in the keynote. We all know Cybe Ops is a huge issue. Cloud is here. You've got industrial IOT going on. And IOT. But these examples that DevNet is showing is that these new capabilities with I won't say a hack but a maker faire culture. >> It is a maker culture right. Which is lot of DIY stuff. So this lot of learning by playing with the API's and multiple one of them. And you'd really find use cases you have never addressed before. We also have a design thinking workshop here going on. And part of it is really thinking about the use cases from the user prospective. What you are trying to address. Before finding the cool technologies. Really understand what your users' needs are. >> Yeah. >> And we are doing a lot of things around that. And bringing it connecting it back to the APIs. Once we learn the right needs. And finding these use cases that were never possible before. >> Well I talked to Susie all the time about this. >> Yeah. >> And I know she's really hardcore on this. But you guys have nailed the community aspect as well. You've brought that open source ethos into the formula. Which makes it more collaborative. No one wants to be alone. I mean the last thing a network engineer wants to do is be the old way of being tied to the chair on the network. Troubleshooting problems. They want to have more collaboration As some of this creativity kicks in. So it's really a new time. How are you guys handling this? Is it like people are having an awakening moment? Or what are you guys doing to nature this? What are some of the exciting things? >> And the best part about the community is that communities learning with each other. Right? It's this feeling of we are enabling our community both traditionally and through even like Cisco Live and DevNet Create. We bring them together to be able to learn from each other much as we learn with them. And trying to define the right use cases and solutions. And that's what the company's behind. The 500,000 developers who are coming and learning with us. They have found the use cases they were addressing for their business. They also found a new skill set that they were looking to learn before. >> Yeah. >> And a lot of them have come along where they are showing their tech cred in the community. Really being the community leaders. >> You know it's been kind of a downer some of the narrative I've seen from press outlets other press outlets and other kind of naysayers has been Hey network guys. You're gonna be automated away. Go learn how to code to save your career. Actually that's not happening. >> That is not happening at all. >> The power of networking certainly as security moves down lower on the stack. And policy and these cool service oriented service meshes. Kubernetes. Really points to the relevance of the network engineer more than ever. You've got SDN. Software Defined Data Center. That's not going away. Automation is going to take mundane tasks away. >> Yes. >> But actions happening at the app layer. >> They have that expertise and 20 years plus experience knowing how networks should be running to make these things possible. The use cases around the applications possible. >> They're more relevant than ever. >> They are more relevant than ever. I would say. Exactly. That's the key. >> Well you guys are at the beginning I think of another set of inflection point. Certainly DevNet's gone in a quick four years. You're connecting to the Cloud Native World with DevNet Create. Which is phenomenal. Those are two worlds that are coming together. I just see another inflection point coming. Maybe it's a million developers. But you've been success in the enterprise where it's been really difficult. Even Microsoft with their legacy developer program .net. The Visual Basic and all the MSDM stuff. >> At the by GitHub >> Yup. >> To kind of maintain relevance. Other companies like Oracle VM wear and other ones they're having a hard time. You guys are just kicking butt. >> So part of it for us is not only focusing on traditional infrastructure. But also talking about the app developer. So these application developers who did not know about network at all. A lot of times they had to fight with their networking juniors to get their application the particular function they wanted to have. Right? So that what we are enabling by bringing them together. Also we have been running small programs like we are trying new markets. Global markets. China, India and some of the things like really reaching out to the big large hackathons. Which are traditionally. For example in India we were recently doing a smart India hackathon. >> Nice. >> There are 500,00 students participated in solving real problems for the country. And DevNet was the provider of applications and API's. Bringing them into the application world with the understanding of network. >> A lot of growth in India and China. Certainly massive new developers coming on board. Okay final question to wrap up the segment. I gotta get your prospective. Take your DevNet hat off for a second. >> Okay. Put your Cisco hat on. >> Sure. For the folks who couldn't make Cisco Live this year what's the big story coming out of the event this year? You guys have been successful with the 500,000 developers. What's the big story developing here? What should people know is the most important story for Cisco Live 2018? >> I think the biggest story I would like to call out is that network is open for business. Network is really open for you to really come and make your intent. Your use cases. Your business outcomes possible. And that's the biggest story I will call out. >> Shubha Govil here product management for DevNet. Here on the Cube. Live coverage. Day two of three days. I'm John Furrier. Stay with us for more live coverage. As we start winding down day two. A lot of great action. The network is programmable. It's creating value and new use cases. And the developers are in the center of the action. The network engineers seeing a clear path of the Cloud and more. We'll be back with more after this short break. (upbeat music)

>> Hello everyone. My name is Zoltan Toroczkai. I am from University of Notre Dame, Physics department, and I'd like to thank the organizers for their kind invitation to participate in this very interesting and promising workshop. Also like to say that I look forward to collaborations with the Redefine Lab and Yoshian collaborators on the topics of this work. So today I'll briefly talk about, our attempt to understand, the fundamental limits of analog, continuous-time computing at least from the point of view of Boolean Satisfiability problem-solving using ordinary differential equations. But I think the issues that we raise during this occasion actually apply to other approaches, analog approaches as well, until to other problems as well. I think everyone here, knows what Boolean Satisfiability problems are. You have N Boolean variables, you have M clauses. Each a disjunction of K literals. Literal is a variable or it's negation. And the goal is to find an assignment to the variable such that all the clauses are true. This is a decision type problem from the NP class, which means you can check in polynomial time for satisfiability of any assignment. And the 3-SAT is NP-complete with K, 3 or larger, which means an efficient 3-SAT solver, (clears throat) implies an efficient solver for all the problems in the NP clause because all the problems in the NP clause can be reduced in polynomial time to 3-SAT. As a matter of fact you can, reduce the NP-complete problems into each other. You can go from 3-SAT to Set Packing or to Maximum Independent Set which is the set packing in graph theoretic notions or terms, to the ising graph SAT problem decision version. This is useful when you are comparing different approaches or working on different kinds of problems. When not all the clauses can be satisfied, you're looking at the optimization version of SAT, called Max-SAT and the goal here is to find the assignment that satisfies the maximum number of clauses, and this is from the NP-hard class. In terms of applications, if we had an efficient SAT solver, or NP-complete problem solver, it would literally, positively influence thousands of problems in applications in industry and science. I'm not going to read this. But this of course gives us some motivation, to work on these kind of problems. Now, our approach to SAT solving, involves embedding the problem in a continuous space, and you use all these to do that. So instead of working zeros and ones, we work with minus one and plus ones, and if we allow the corresponding variables, to change continuously between the two bounds, we formulate the problem with the help of a Clause Matrix. If, if a clause does not contain a variable or its negation, the corresponding matrix element is zero. If it contains the variable in positive form it's one. If it contains the variable in negated form, it's negative one. And now we use this to formulate these products, called clause violation functions, one for every clause, which rarely continues between zero and one and beyond zero if and only if the clause itself is true. Then we form... We define, also define the dynamics, search dynamics in this and the M-dimensional hypercube, where the search happens and if there exists solutions they're sitting in some of the corners of this hypercube. So we define this energy, potential or landscape function as shown here in a way that it, this is zero if and only if all the clauses, all the Kms are zero. All the clauses are satisfied, keeping these auxiliary variables, Ams always positive. And therefore what we do here is a dynamics that is essentially a gradient descent on this potential energy landscape. If you are to keep all the Ams constant then it would get stuck in some local minimum. However what do you do here is, we couple it with the dynamics. We couple it with the clause violation functions as shown here. And if you didn't have these Am here, just had just the Kms, for example, you have essentially, both case you have a positive feedback. You have a decreasing variable, but in that case you'll still get stuck, would still behave... We'll still find solutions better than the constant version or still would get stuck. Only when we put here this Am, which makes them dynamics in this variable exponential like, only then it keeps searching until it finds a solution. And there's a reason for that, that I'm not going to talk about here, but essentially boils down to performing a gradient descent on a globally time-varying landscape. And, and, and this is what works. Now, I'm going to talk about the good or bad, and maybe the ugly. This is, this is... What's good is that it's a hyperbolic dynamical system, which means that if you take any domain in the search space that doesn't have a solution in it or any solution in it, then the number of trajectories in it, the case exponentially quickly and the decay rate is a characteristic, invariant characteristic of the dynamics itself with the dynamical systems called the escape rate. The inverse of that is the timescale in which you find solutions by this dynamical system. And you can see here some trajectories, they are curved because it's, it's not linear but it's transiently curved to give, if there are solutions of course, we could see eventually, it does lead to the solutions. Now, in terms of performance, here what you show, for a bunch of, constraint densities, defined by, M over N, the ratio between clauses to variables, for random SAT problems, is random 3-SAT problems. And they, they, as, as function of N, and we look at, monitor the wall time, the wall clock time, and it, it behaves quite well, it behaves as a, as a polynomialy, until you actually hit, or reach the set on set transition, where the hardest problems are found. But what's more interesting is if you monitor the continuous-time t, the performance in terms of the analog continuous-time t, because that seems to be a polynomial. And the way we show that, is we can see the random K-SAT or random 3-SAT for a fixed constraint density. And we here, what you show here is at the, right at the threshold where it's really hard. And, (clears throat) we monitor the fraction of problems that we have not been able to solve it. We select thousands of problems at that cost rate ratio and we solve them with our algorithm, and we monitor the fraction of problems that have not yet been solved by continuous-time t. And these, as you see these decays exponentially in different decay rates for different system sizes and in this spot shows that this decay rate behaves polynomialy. or actually as a power law. So if you combine these two, you find that the time needed to solve all problems, except maybe appeared fraction of them, scales polynomialy with problem size. So you have polynomial continuous-time complexity. And this is also true, for other types of very hard constraints of the SAT problem such as exact color, because you can always transform them into 3-SAT as we discussed before, Ramsay coloring and, and on these problems, even algorithms like a survey propagation wheel will fail. But this doesn't mean that P equals NP because what you have, first of all, if you were to implement these equations in a device, whose behavior is described by these ODEs, then of course, t the continuous-time variable, becomes a physical wall clock time. And that would be polynomialy scaling but you have other variables, auxiliary variables, which fluctuate in an exponential manner. So if they represent currents or voltages in your realization and it would be an exponential cost algorithm. But this is some kind of trade between time and energy while I know how to generate energy or I don't know how to generate time but I know how to generate energy so it could be useful. But there's other issues as well, especially if you're trying to do this on a digital machine, but also happens, problems happen, appear, other problems appear on in physical devices as well as we discuss later. So if you implement these in GPU, you can, then you can get an order of two magnitude speedup, and you can also modify this, to solve Max-SAT problems quite efficiently, we are competitive with the best, heuristics solvers, this is all the problems in 2016, Max-SAT competition. So, so this, this, this is definitely, this is like a good approach, but there's of course, interesting limitations, I would say interesting, because it kind of makes you think about what it needs and how you can explore this, these observations in understanding better analog continuous-time complexity. If you monitor the discrete number, the number of discrete steps, done by the Runge Kutta integrator, and you solve this on a digital machine. You're using some kind of integrator, and, you know, using the same approach, but now you measure the number of problems you haven't solved, by a given number of discrete steps taken by the integrator. You find out, you have exponential discrete-time complexity. And of course, this is a problem. And if you look closely, what happens, even though the analog mathematical trajectory, that's the red curve here, if you monitor what happens in discrete time, the integrator fluctuates very little. So this is like you know, third or four digits precision, but fluctuates like crazy. So it really is like the integration freezes out, and this is because of the phenomenon of stiffness that I'll talk a little bit, more about a little bit later. So you know, it may look like an integration issue on your digital machines that you could improve and you could definitely improve, but actually the issue is bigger than that. It's, it's deeper than that because on a digital machine there is no time energy conversion. So the auxiliary variables are efficiently represented in a digital machine, so there's no exponential fluctuating current or voltage in your computer when you do this. So if e is not equal NP, then the exponential time complexity or exponential cost complexity has to hit you somewhere. And this is how. But you know one would be tempted to think maybe, this wouldn't be an issue in a analog device, and to some extent is true. Analog devices can be orders of magnitude faster, but they also suffer from their own problems because P not equal NP affects that clause of followers as well. So, indeed if you look at other systems, like Coherent Ising Machine with Measurement-Feedback, or Polariton Condensate Graphs or Oscillator Networks, they all hinge on some kind of, our ability to control real variables with arbitrarily high precision, and Oscillator Networks, you want to read out arbitrarily close frequencies. In case of CIMs, we require identical analog amplitudes which is hard to keep and they kind of fluctuate away from one another, shift away from one another, And, and if you control that, of course, then you can control the performance. So, actually one can ask if whether or not this is a universal bottleneck, and it seems so, as I will argue next. We can recall a fundamental result by A. Schönhage, Graham Schönhage from 1978 who says that, it's a purely computer science proof, that, "If you are able to compute, "the addition, multiplication, division "of real variables with infinite precision then, "you could solve NP-complete problems in polynomial time." He doesn't actually propose a solid work, he just shows mathematically that this will be the case. Now, of course, in real world, you have loss of precision. So the next question is, "How does that affect the computation of our problems?" This is what we are after. Loss of precision means information loss or entropy production. So what we are really looking at, the relationship between hardness and cost of computing of a problem. (clears throat) And according to Sean Harget, there is this left branch, which in principle could be polynomial time, but the question, whether or not this is achievable, that is not achievable, but something more achievable that's on the right-hand side. You know, there's always going to be some information loss, some entropy generation that could keep you away from, possibly from polynomial time. So this is what we'd like to understand. And this information loss, the source of this is not just noise, as, as I will argue in any physical system, but it's also of algorithmic nature. So that is a questionable area or, or approach, but Schönhage's result is purely theoretical, no actual solver is proposed. So we can ask, you know, just theoretically, out of curiosity, "Would in principle be such solvers?" Because he's not proposing a solver. In such properties in principle, if you were to look mathematically, precisely what that solver does, would have the right properties. And I argue, yes, I don't have a mathematical proof but I have some arguments that this would be the case. And this is the case for actually our sitdia solver, that if you could calculate, it's subjectivity in a loss this way, then it would be... Would solve NP-complete problems in polynomial continuous-time. Now, as a matter of fact, this is a bit more difficult question because time in all these can be re-scaled however you want. So what Bournez says, that you actually have to measure the length of the trajectory which is an invariant of the dynamical system or the property of the dynamical system, not of it's parametrization. And we did that. So Shubha Kharel my student did that, by first improving on the stiffness of the problem of the integrations using the implicit solvers and some smart tricks, such that you actually are closer to the actual trajectory and using the same approach to know, what fraction of problems you can solve. We did not give a length of the trajectory, you find that it is polynomialy scaling with the problem size. So we have polynomial scale complexity. That means that our solver is both poly-length, and as it is defined, it's also poly-time analog solver. But if you look at as a discrete algorithm, which will measure the discrete steps on a digital machine, it is an exponential solver, and the reason is because of all this stiffness. So every integrator has to truncate, digitize and truncate the equations. And what it has to do is to keep the integration within this so-called Stimpy TD gen for, for that scheme. And you have to keep this product within Eigenvalues of the Jacobian and the step size within this region, if you use explicit methods, you want to stay within this region. But what happens, that some of the eigenvalues grow fast for stiff problems, and then you're, you're forced to reduce that t, so the product stays in this bounded domain, which means that now you have to, we are forced to take smaller and smaller time steps, so you're, you're freezing out the integration and what I will show you, that's the case. Now you can move to implicit solvers, which is a new trick, in this case, your stability domain is actually on the outside, but what happens in this case, is some of the eigenvalues of the Jacobian, also for this instant start to move to zero, as they are moving to zero, they are going to enter this instability region. So your solver is going to try to keep it out, so it's going to increase the delta t, but if you increase that t, you increase the truncation errors, so you get randomized in the large search space. So it's, it's really not, not willing to work out. Now, one can sort of, introduce a theory or a language to discuss computational, analog computational complexity, using the language from dynamical systems theory. But basically I don't have time to go into this but you have for hard problems, the chaotic object the chaotic saddle in the middle of the search space somewhere, and that dictates how the dynamics happens and invariant properties of the dynamics, of course, of that saddle is what determines performance and many things. So an important measure that we find that, is also helpful in describing, this analog complexity is the so-called Kolmogorov or metric entropy. And basically what this does in an intuitive way, is to describe the rate at which the uncertainty, containing the insignificant digits of a trajectory in the back, they flow towards the significant ones, as you lose information because of errors being, grown or, or or, or developed into larger errors in an exponential, at an exponential rate because you have positive Lyapunov exponents. But this is an invariant property. It's the property of the set of all these, not how you compute them. And it's really the intrinsic rate of accuracy loss of a dynamical system. As I said that you have in such a high dimensional dynamical system, you have positive and negative Lyapunov exponents, as many as the total is the dimension of the space and user dimension, the number of unstable manufactured dimensions and assets now more stable many forms dimensions. And there's an interesting and I think important Pesin equality, equality called the Pesin equality, that connects the information theoretic, as per the rate of information loss with the geometric data each trajectory separate minus cut part which is the escape rate that I already talked about. Now, one can actually prove a simple theorem strike back of the calculation. The idea here is that, you know the rate at which the largest rate at which the closely started trajectory, separate from one another. So now you can say that, that is fine, as long as my trajectory finds the solution, before the trajectory separate too quickly. In that case, I can have the hope, that if I start from some region of the face space, several closely started trajectories, they kind of go into the same solution over time and that's, that's, that's this upper bound of this limit. And it is really showing that it has to be... It's an exponentially smaller number, but it depends on the N, dependence of the exponent right here, which combines information loss rate and the solution time performance. So these, if these exponent here or there, has a large independence, so even a linear independence, then you really have to start trajectories, exponentially closer to one another, in order to end up in the same order. So this is sort of like the, the direction that you are going into, and this formulation is applicable to, to all dynamical systems, deterministic dynamical systems. (clears throat) And I think we can expand this further because the, there is a way of getting the expression for the escape rates in terms of N the number of variables from cycle expansions, that I don't have time to talk about, but it's kind of like a program that you can try to pursue. And this is it. So uh, uh... The conclusions, I think are self-explanatory. I think there is a lot of future in, in analog continuous-time computing. They can be efficient by orders of magnitude than digital ones in solving NP-hard problems, because first of all, many of the systems lack of von Neumann bottleneck, there's parallelism involved and you can also have a larger spectrum of continuous-time dynamical algorithms than discrete ones. And, and, you know, but we also have to be mindful of what are the possibilities, what are the limits? And one, one open question, if any important open question is you know, "What are these limits? "Is there some kind of no-go theorem that tells you that, "you can never perform better than this limit "or, or that limit?" And I think that's, that's the exciting part to, to derive these, these limits and to get to an understanding about what's possible in this, in this area. Thank you.