(upbeat music) >> Announcer: theCUBE presents KubeCon and CloudNativeCon Europe 2022, brought to you by Red Hat, the Cloud Native Computing Foundation, and its ecosystem partners. >> Welcome to Valencia, Spain, in KubeCon CloudNativeCon Europe 2022. I'm your host Keith Townsend, along with Paul Gillon. And we're going to talk to some amazing folks. But first Paul, do you remember your college days? >> Vaguely. (Keith laughing) A lot of them are lost. >> I think a lot of mine are lost as well. Well, not really, I got my degree as an adult, so they're not that far past. I can remember 'cause I have the student debt to prove it. (both laughing) Along with us today is Kenneth Hoste, systems administrator at Ghent University, and Marcel Hild, senior manager software engineering at Red Hat. You're working in office of the CTO? >> That's absolutely correct, yes >> So first off, I'm going to start off with you Kenneth. Tell us a little bit about the research that the university does. Like what's the end result? >> Oh, wow, that's a good question. So the research we do at university and again, is very broad. We have bioinformaticians, physicists, people looking at financial data, all kinds of stuff. And the end result can be very varied as well. Very often it's research papers, or spinoffs from the university. Yeah, depending on the domain I would say, it depends a lot on. >> So that sounds like the perfect environment for cloud native. Like the infrastructure that's completely flexible, that researchers can come and have a standard way of interacting, each team just use it's resources as they would, the Navana for cloud native. >> Yeah. >> But somehow, I'm going to guess HPC isn't quite there yet. >> Yeah, not really, no. So, HPC is a bit, let's say slow into adopting new technologies. And we're definitely seeing some impact from cloud, especially things like containers and Kubernetes, or we're starting to hear these things in HPC community as well. But I haven't seen a lot of HPC clusters who are really fully cloud native. Not yet at least. Maybe this is coming. And if I'm walking around here at KubeCon, I can definitely, I'm being convinced that it's coming. So whether we like it or not we're probably going to have to start worrying about stuff like this. But we're still, let's say, the most prominent technologies of things like NPI, which has been there for 20, 30 years. The Fortran programming language is still the main language, if you're looking at compute time being spent on supercomputers, over 1/2 of the time spent is in Fortran code essentially. >> Keith: Wow. >> So either the application itself where the simulations are being done is implemented in Fortran, or the libraries that we are talking to from Python for example, for doing heavy duty computations, that backend library is implemented in Fortran. So if you take all of that into account, easily over 1/2 of the time is spent in Fortran code. >> So is this because the libraries don't migrate easily to, distributed to that environment? >> Well, it's multiple things. So first of all, Fortran is very well suited for implementing these type of things. >> Paul: Right. >> We haven't really seen a better alternative maybe. And also it'll be a huge effort to re-implement that same functionality in a newer language. So, the use case has to be very convincing, there has to be a very good reason why you would move away from Fortran. And, at least the HPC community hasn't seen that reason yet. >> So in theory, and right now we're talking about the theory and then what it takes to get to the future. In theory, I can take that Fortran code put it in a compiler that runs in a container? >> Yeah, of course, yeah. >> Why isn't it that simple? >> I guess because traditionally HPC is very slow at adopting new stuff. So, I'm not saying there isn't a reason that we should start looking at these things. Flexibility is a very important one. For a lot of researchers, their compute needs are very picky. So they're doing research, they have an idea, they want you to run lots of simulations, get the results, but then they're silent for a long time writing the paper, or thinking about how to, what they can learn from the results. So there's lots of peaks, and that's a very good fit for a cloud environment. I guess at the scale of university you have enough diversity end users that all those peaks never fall at the same time. So if you have your big own infrastructure you can still fill it up quite easily and keep your users happy. But this busty thing, I guess we're seeing that more and more or so. >> So Marcel, talk to us about, Red Hat needing to service these types of end users. That it can be on both ends I'd imagine that you have some people still in writing in Fortran, you have some people that's asking you for objects based storage. Where's Fortran, I'm sorry, not Fortran, but where is Red Hat in providing the underlay and the capabilities for the HPC and AI community? >> Yeah. So, I think if you look at the user base that we're looking at, it's on this spectrum from development to production. So putting AI workloads into production, it's an interesting challenge but it's easier to solve, and it has been solved to some extent, than the development cycle. So what we're looking at in Kenneth's domain it's more like the end user, the data scientist, developing code, and doing these experiments. Putting them into production is that's where containers live and thrive. You can containerize your model, you containerize your workload, you deploy it into your OpenShift Kubernetes cluster, done, you monitor it, done. So the software developments and the SRE, the ops part, done, but how do I get the data scientist into this cloud native age where he's not developing on his laptop or on a machine, where he SSH into and then does some stuff there. And then some system admin comes and needs to tweak it because it's running out of memory or whatnot. But how do we take him and make him, well, and provide him an environment that is good enough to work in, in the browser, and then with IDE, where the workload of doing the computation and the experimentation is repeatable, so that the environment is always the same, it's reliable, so it's always up and running. It doesn't consume resources, although it's up and running. Where it's, where the supply chain and the configuration of... And the, well, the modules that are brought into the system are also reliable. So all these problems that we solved in the traditional software development world, now have to transition into the data science and HPC world, where the problems are similar, but yeah, it's different sets. It's more or less, also a huge educational problem and transitioning the tools over into that is something... >> Well, is this mostly a technical issue or is this a cultural issue? I mean, are HPC workloads that different from more conventional OLTP workloads that they would not adapt well to a distributed containerized environment? >> I think it's both. So, on one hand it's the cultural issue because you have two different communities, everybody is reinventing the wheel, everybody is some sort of siloed. So they think, okay, what we've done for 30 years now we, there's no need to change it. And they, so it's, that's what thrives and here at KubeCon where you have different communities coming together, okay, this is how you solved the problem, maybe this applies also to our problem. But it's also the, well, the tooling, which is bound to a machine, which is bound to an HPC computer, which is architecturally different than a distributed environment where you would treat your containers as kettle, and as something that you can replace, right? And the HPC community usually builds up huge machines, and these are like the gray machines. So it's also technical bit of moving it to this age. >> So the massively parallel nature of HPC workloads you're saying Kubernetes has not yet been adapted to that? >> Well, I think that parallelism works great. It's just a matter of moving that out from an HPC computer into the scale out factor of a Kubernetes cloud that elastically scales out. Whereas the traditional HPC computer, I think, and Kenneth can correct me here is, more like, I have this massive computer with 1 million cores or whatnot, and now use it. And I can use my time slice, and book my time slice there. Whereas this a Kubernetes example the concept is more like, I have 1000 cores and I declare something into it and scale it up and down based on the needs. >> So, Kenneth, this is where you talked about the culture part of the changes that need to be happening. And quite frankly, the computer is a tool, it's a tool to get to the answer. And if that tool is working, if I have a 1000 cores on a single HPC thing, and you're telling me, well, I can't get to a system with 2000 cores. And if you containerized your process and move it over then maybe I'll get to the answer 50% faster maybe I'm not that... Someone has to make that decision. How important is it to get people involved in these types of communities from a researcher? 'Cause research is very tight-knit community to have these conversations and help that see move happen. >> I think it's very important to that community should, let's say, the cloud community, HPC research community, they should be talking a lot more, there should be way more cross pollination than there is today. I'm actually, I'm happy that I've seen HPC mentioned at booths and talks quite often here at KubeCon, I wasn't really expecting that. And I'm not sure, it's my first KubeCon, so I don't know, but I think that's kind of new, it's pretty recent. If you're going to the HPC community conferences there containers have been there for a couple of years now, something like Kubernetes is still a bit new. But just this morning there was a keynote by a guy from CERN, who was explaining, they're basically slowly moving towards Kubernetes even for their HPC clusters as well. And he's seeing that as the future because all the flexibility it gives you and you can basically hide all that from the end user, from the researcher. They don't really have to know that they're running on top of Kubernetes. They shouldn't care. Like you said, to them it's just a tool, and they care about if the tool works, they can get their answers and that's what they want to do. How that's actually being done in the background they don't really care. >> So talk to me about the AI side of the equation, because when I talk to people doing AI, they're on the other end of the spectrum. What are some of the benefits they're seeing from containerization? >> I think it's the reproducibility of experiments. So, and data scientists are, they're data scientists and they do research. So they care about their experiment. And maybe they also care about putting the model into production. But, I think from a geeky perspective they are more interested in finding the next model, finding the next solution. So they do an experiment, and they're done with it, and then maybe it's going to production. So how do I repeat that experiment in a year from now, so that I can build on top of it? And a container I think is the best solution to wrap something with its dependency, like freeze it, maybe even with the data, store it away, and then come to it back later and redo the experiment or share the experiment with some of my fellow researchers, so that they don't have to go through the process of setting up an equivalent environment on their machines, be it their laptop, via their cloud environment. So you go to the internet, download something doesn't work, container works. >> Well, you said something that really intrigues me you know in concept, I can have a, let's say a one terabyte data set, have a experiment associated with that. Take a snapshot of that somehow, I don't know how, take a snapshot of that and then share it with the rest of the community and then continue my work. >> Marcel: Yeah. >> And then we can stop back and compare notes. Where are we at in a maturity scale? Like, what are some of the pitfalls or challenges customers should be looking out for? >> I think you actually said it right there, how do I snapshot a terabyte of data? It's, that's... >> It's a terabyte of data. (both conversing) >> It's a bit of a challenge. And if you snapshot it, you have two terabytes of data or you just snapshot the, like and get you to do a, okay, this is currently where we're at. So that's why the technology is evolving. How do we do source control management for data? How do we license data? How do we make sure that the data is unbiased, et cetera? So that's going more into the AI side of things. But at dealing with data in a declarative way in a containerized way, I think that's where currently a lot of innovation is happening. >> What do you mean by dealing with data in a declarative way? >> If I'm saying I run this experiment based on this data set and I'm running this other experiment based on this other data set, and I as the researcher don't care where the data is stored, I care that the data is accessible. And so I might declare, this is the process that I put on my data, like a data processing pipeline. These are the steps that it's going through. And eventually it will have gone through this process and I can work with my data. Pretty much like applying the concept of pipelines through data. Like you have these data pipelines and then now you have cube flow pipelines as one solution to apply the pipeline concept, to well, managing your data. >> Given the stateless nature of containers, is that an impediment to HPC adoption because of the very large data sets that are typically involved? >> I think it is if you have terabytes of data. Just, you have to get it to the place where the computation will happen, right? And just uploading that into the cloud is already a challenge. If you have the data sitting there on a supercomputer and maybe it was sitting there for two years, you probably don't care. And typically a lot of universities the researchers don't necessarily pay for the compute time they use. Like, this is also... At least in Ghent that's the case, it's centrally funded, which means, the researchers don't have to worry about the cost, they just get access to the supercomputer. If they need two terabytes of data, they get that space and they can park it on the system for years, no problem. If they need 200 terabytes of data, that's absolutely fine. >> But the university cares about the cost? >> The university cares about the cost, but they want to enable the researchers to do the research that they want to do. >> Right. >> And we always tell researchers don't feel constrained about things like compute power, storage space. If you're doing smaller research, because you're feeling constrained, you have to tell us, and we will just expand our storage system and buy a new cluster. >> Paul: Wonderful. >> So you, to enable your research. >> It's a nice environment to be in. I think this might be a Jevons paradox problem, you give researchers this capability you might, you're going to see some amazing things. Well, now the people are snapshoting, one, two, three, four, five, different versions of a one terabytes of data. It's a good problem to have, and I hope to have you back on theCUBE, talking about how Red Hat and Ghent have solved those problems. Thank you so much for joining theCUBE. From Valencia, Spain, I'm Keith Townsend along with Paul Gillon. And you're watching theCUBE, the leader in high tech coverage. (upbeat music)

>>Welcome everybody to the cubes live coverage of AWS reinvent 2021. We're here in the main hall. Yes, this is a physical event. It's a hybrid event, probably the industry's most important hybrid event in the year. We're super excited to be here. Of course, last year during the lockdown, reinvent was purely virtual. This year. They go in hybrid 20 plus thousand people. I hear the whisper numbers like 25, 20 7,000 hundreds of thousands of people online. The cubes here, two sets, we've got two remote studios, super excited. I'd like to introduce my co-host David Nicholson. He'll be here all week with us. Uh, John furrier is also here, Lisa Martin for the cubes wall-to-wall coverage. And we're so psyched to start off this session with Kenneth Chestnut. Who's the head of technology partnerships at Stripe. Stripe's an amazing company, Ken. Great to see you. Thanks for coming on. >>Thanks for having me, Dave and David. I greatly appreciate it. How about this? >>Right. Finally live event. We've done a few. We probably done four or five this year, but >>It's good to be back in person. It is. Yeah, absolutely. It's >>A Stripe. I mean, wow. Can a powering the new economy. Tell us a little bit more for those people who may not be familiar with Stripe. They probably use it without even knowing it when they sign it away. Yeah. So tell us about the >>Well, uh, Stripe was founded in 2010 by two brothers, Patrick and John Colson. And really it was from their first business and realizing how hard it was to actually charge for things on online. Um, you had to acquire a relationship with, uh, with a gateway provider to accept payments. You had to acquire a relationship with a, with a acquiring bank. Um, and you had to do that for each and every country that you wanted to service. Uh, so the same way that AWS reduced the barrier in terms of not having to procure, spend millions of dollars on storage, computers, networking, uh, effectively, what we we've done at Stripe is reduce the barriers around economic infrastructure, accepting payments online, >>Use that undifferentiated heavy lifting for payments. So describe Ken, what it was like kind of pre Stripe. You would literally have to install servers, get storage and put, put software on there, get a database. And then what if you had any money left over, you can actually do some business, but, but describe the sort of what the experience is like with Stripe. >>Sure. So, uh, the R R with, with Stripe, we literally talk about seven lines of code. So we, we allow any developer to, um, uh, provide a set of APIs for any developer to accept payments on online. And we do the undifferentiated heavy lifting in terms of accepting payments, accepting those payments, processing them revenue, reporting, and reconciliation, um, all ensuring compliance and security. Um, so it's like you said, uh, taking care of the undifferentiated heavy lifting are around accepting payments online in the enabling >>The enabler. There is the cloud. I mean, it was 2009, 2010. You guys were founded, the cloud was only like three years old. Right. And so you had to really sort of take a chance on leveraging the cloud or maybe early on you just installed it yourself and said, this isn't going to scale. So maybe tell us how you sort of leverage the cloud. >>Sure. Um, so we're a long time, uh, AWS, uh, customer and user, um, uh, back in the early days of, of Stripe in the early days of, of AWS. And we've just grown, uh, with, with AWS and the ecosystem. And it's interesting because a lot of, uh, a lot of the companies that have been built on, on AWS and grown to be successful, they're also Stripe customers as well. So they use Stripe for their economic infrastructure. >>We use Stripe, we run our company on AWS and we use Stripe. It it's true. The integration took like minutes. It was so simple. Hey it, test it, make sure it scales. But so what, what's the stack look like? What is there, is there such thing as a payment stack? What's the technology stuff? >>Sure. So we initially started with payments and being able to accept payments, uh, on online. Uh we've we brought in out our, our, our Stripe product portfolio now to effectively provide economic, uh, infrastructure for the internet. So that could be accepting payments. Uh, it could be setting up marketplaces. So companies like Lyft and Deliveroo, uh, use Stripe to power their marketplaces with their, with their drivers and, and, um, uh, delivers, um, uh, we provide, uh, a product called radar that, uh, that, um, prevents fraud, uh, around, around the globe. Um, based upon the data that we're seeing from our, from our customers, um, we have, uh, issuing and treasury so that companies can provide their users or their merchants with banking services. So loans, uh, issuing credit cards. So we we've really broadened out the product portfolio of Stripe to provide sort of economic infrastructure for the internet. So >>We talked about strike being in the cloud from an infrastructure perspective and how that enables certain things, but that in and of itself, doesn't change the dynamics around sovereignty and governance from country to country. Sure. Uh, I imagine that the global nature of AWS sort of dovetails with your strategy, but how, how do you address that? It's one thing to tell me in Northern California, you can process payments for me, but now globally go across 150 countries. How do you make that work? Yeah, >>Uh, absolutely. So we, we establish relationships, uh, within, within each company country that we operate in we're in about 47, uh, countries, uh, today, um, and that's rapidly expanding so that companies can, can process or accept payments and do, uh, financial transactions within, within, within those countries. So we're in 47 countries today. We, we accept a multitude of different payment, uh, different currencies, different payment types. So the U S is very, uh, credit card focused. But if you go to other, other parts of the globe, it could be a debit cards. It could be, um, uh, wallets, uh, uh, Google pay, Ali pay, uh, others. So really it's, uh, providing sort of the payment methods that users prefer in, in the different countries, uh, and meeting and meeting those users where, where they are. >>Are you out of the box compliant? What integration is required to do that? Uh, what about things like data sovereignty, is that taken care of by the cloud provider or you guys, and where, w w where does, where does AWS end and you guys pick up? Yes, >>We're, we're PCI compliant. Um, we, we leverage AWS as our, as our infrastructure, um, to grow, grow and scale. So, um, one of the things that we're, we're proud of is, uh, through, throughout 2020 and 2021, we've, we've had 11 nines of, uh, of, of, uh, or five nines of uptime, um, even through, um, uh, black Friday and cyber Monday. So providing AWS provides that, that infrastructure, which we built on top of to provide, uh, you know, five nines of uptime for our, for our users. >>You describe in more detail, Kenya, your ecosystem. I mean, you're responsible for tech partnerships. What does that ecosystem, how I paint a picture of it? >>Sure. So, um, uh, a number of users want to be able to use Stripe with, with their other, uh, it infrastructure and, and their business processes. So a customer may start, uh, with a salesperson may start with a quote or order, uh, in, in Salesforce, want to automate the invoicing and billing and payment of that with, with Stripe and then, uh, reconcile re revenue and an ERP solution like SAP or Oracle or NetSuite or into it, um, in the case of, of small, medium businesses. So really, um, what we're focused on is building out that, that ecosystem to allow, uh, um, our, our customers to streamline their business processes, um, and, and integrate Stripe into their existing it infrastructure and, and business processes. >>You mentioned a lot of different services, but broadly speaking, if I think about payments, correct me if I'm wrong, but you were one of the early, uh, sort of software companies, if I can call you that, um, platforms, whatever, but to really focus on a usage based pricing, but how do I, how do I engage with you? What's, what's the pricing model. Maybe you could describe that a little. >>Sure. So the pricing model is very, very transparent. Uh, it's on, it's on the website. So, uh, we, we take a, um, a percentage of each transaction. So literally you can, you can set up a, a Stripe account it's self-service, um, uh, we, we take a 2.9% plus 30 cents on every, uh, Tran transaction. Um, we don't, you don't start getting, um, uh, charged until, uh, you start accepting payments from your, from your customers or from your users. >>Um, can you give us a sense of the business scope, maybe any metrics you can share, customers, whatever. >>Sure. So there's a couple of things we can share publicly, just in terms of the size of the business. I think since, uh, since 2020, uh, more than 2 million businesses have launched on, on Stripe. Uh, so, uh, 2 million in, in, in, in 2020, um, we've, uh, uh, in the past 12 months, we've, uh, uh, uh, processed over 173 billion, uh, API calls. Uh, we do we process about, um, uh, hundreds of billions of, of, of, uh, payment volume, uh, every, every year. Um, if you look at sort of the macros of the business, the business is growing faster than the broader e-commerce space. So the amount of payment volume that we did in this past year is more than the entire industry did when Patrick and John founded the company. And in 2010, just to give you a, uh, an idea of the, the, the size of the business and sort of the pace of the business >>You're growing as e-commerce grows, but you're also stealing share from other sort of traditional payment systems. Okay. So that's a nice flywheel effect. And of course, Stripe's a private company they've raised well over a billion dollars of Peter teal, and it wasn't original founders, so are funders. So, you know, that's, he's talking scale. I want to go back to something you said about radar. Sure. So there's tech in your stack fraud detection, right. So some of >>That in machine learning, right. >>So, and so you guys, I mean, are you a technology company, are you a F a FinTech company? What are you? >>We're a software company. We provide software and we provide technology for developers, uh, to make online businesses and make, uh, uh, commerce, uh, more seamless and more frictionless >>Cloud-first API first. I mean, maybe describe how that is different maybe than, you know, the technical debt that's been built up over, you know, decades with traditional payment systems. >>Yes, it's very similar to the early, earlier days of AWS where a lot of tech forward companies leveraged Stripe, um, to, um, whether it be large enterprises to transform their businesses and move online, or, or, uh, uh, startups and developers that want to, uh, start a new business online and, and do that, uh, as quickly and seamlessly as possible. So it's, it's quite the gamut from large enterprises that are digitally transforming themselves companies like Marske and, and NASDAQ and others, as well as, uh, um, startups and developers that have started their businesses and born on born on Stripe. So >>When you talk about a startup, how small of an entity makes sense, uh, when you think of, if you look at, from an economic perspective, lowering the friction associated with transactions can lift up a large part of the world with sort of, you know, w with very, very small businesses. Is that something that this is all about? >>Yeah, absolutely. So, like I said, you know, two, 2 million business have sub launched on, on, on Stripe, uh, in, in the past year. And, and those businesses vary, but it could be literally a, a developer or a, uh, uh, a small, uh, SMB that wants to be able to accept payments on online. And it can just set up a Stripe account and start accepting payments. >>Yeah. So this is not a one hit wonder, um, lay out the vision for Stripe, right? I mean, you're, you're a platform, uh, you're, you're becoming a fundamental ingredient of the digital economy sounds pre pandemic. That was all a bunch of buzzwords, but today we all know how important that is, but what lay out the vision for us can, >>Yeah, it really are. The mission of Stripe is to grow the GDP of the internet. Um, and, and so what that means is, uh, more and more our, our, our basic belief is more and more and more businesses, uh, will, will, uh, go, go online, uh, with, uh, with the pandemic that that was, uh, accelerated. But I think that the general trend of businesses moving online, uh, will continue to accelerate, and we want to provide, uh, economic infrastructure to support those businesses. Um, you know, um, uh, uh, Andreessen talked about sort of software, software eating the world well fit. Our belief has FinTech is eating software. So in, in the fullness of time, I think the opportunity is for, uh, any, any company to be a financial services company. And we want to empower any company that wants to, or any user that wants to be a financial services company to, to provide the economic infrastructure for them to do so. >>And, and, you know, I mean your data company in that sense, you're moving bits around, you know, and those datas, I like to say data's eating software, you know, cause really you gotta have your data act together. Absolutely. And that's an evolving, I mean, you guys started to, to 2010, I would imagine your data strategy has evolved quite dramatically. Yeah. >>It's a great, it's a great call out Dave. Uh, one of our other products is a product called Sigma. So Sigma allows, uh, merchants or our customers to query payment and transaction data. So they want to be able to understand who, who, who are their customers, what are the payment methods that those customers prefer in different countries, in different regions? Um, so we're, we're starting to have some interesting use cases, um, working with, with AWS and other partners when you can start combining payment and transaction data in Stripe with other data to understand customer segmentation, customer 360 lifetime value of a customer customer acquisition costs, being able to close the books faster in your ERP, because you can apply that payment and transaction data to your general ledger to, to close the books faster at the end of the month or at the end of the, at the end of the year. So, uh, yeah, we we're, um, uh, as, as more and more companies are using Stripe, um, they want to be able to take advantage of that data and combine it with other, other sources of data to drive business. >>Yeah. You mentioned some of those key metrics that are, that are so important to companies today. I'll give you the last word re-invent this hall is packed, um, a little bit surprising, frankly, you know, but, uh, but exciting. Uh, what are you looking forward to this? >>Yeah, I'm just looking forward to meeting people in person again, it's, uh, it's great to be here and, and, you know, uh, uh, we have a strong relationship with AWS. We have lots of partners in, in, in common here, uh, as well, both consulting partners and technology partners. So really looking forward to meeting with partners and customers, and especially as we, as we plan for next year and, uh, launching our, our, our partner program beginning of next year. Uh, there's a lot of, uh, uh, groundwork and things to learn from, from here. As we, as we, we, we, we launch our, our, our partner business formula next >>I'll bet. Looking forward to that, Ken, thanks so much for coming to the cure. You so much. It was great to have a chat at the time. All right. And we want to thank our sponsors, uh, AWS, of course, and also AMD who's making the editorial segments that we bring you this week possible for Dave Nicholson. I'm Dave Volante. You're watching the cube at AWS reinvent 2021. Keep it right there, right back.

>> From Mountain View, California, it's theCUBE covering the 15th annual Grow Awards. Brought to you by ACG SV. >> Hey, Lisa Martin, on the ground with theCUBE at the 15th annual ACG SV Grow Awards, Association for Corporate Growth Silicon Valley, is what that stands for. Can you hear the energy and the innovation going on back here? It's amazing tonight. I'm very pleased to welcome to theCUBE, one of tonight's winners from Arista Technologies Kenneth Duda, the CTO, SVP of software engineering, and one of the founders of Arista Technologies. Kenneth, thank you so much and congratulations! >> Thank you so much, we're honored by the award. >> Well, it's been amazing. Outstanding Growth Award winner, congratulations. I was just looking at some of the recent earnings from Arista, nice Q4 earnings from FY-18. >> Thank you. >> Above the guidance, stock price rising this year. Last month Goldman Sachs added Arista to its conviction buy list. You guys are on nice trajectory, tell me about that. >> Well, it's just been a fantastic journey, you just don't get this many chances to participate in something like Arista from the ground up. Our growth has been driven in no small part thanks to the incredible growth of cloud computing. Cloud computing is changing the world and the cloud data centers need a different kind of network infrastructure. They need something that scales, meet their needs, and is customizable to integrate with all of their management systems, automation, and we've been able to provide that and be part of that journey, it's been incredibly gratifying. >> So you specifically talk with customers a lot, I was reading about one of your recent big wins in Canada, CBC Radio Canada facility in Montreal, but talk to me about what's some of the things now that you're hearing from customers especially those customers who are still in the process of transforming and transitioning workloads to the cloud. What are some of the things that surprise you about where customers are in any industry in this journey. >> Right, well, so I spend most of my time talking to the enterprise customers because there are so many of them and what we've learned there is a couple of things. One is they are very impacted by cloud. Cloud's a big deal, they're moving somewhere closer to the cloud, they're also building their own internal environments in a more cloud-like fashion and, as such, benefit from Arista's approach. But the most interesting thing I've learned is that neither of those is the most important thing. The most important thing is the network has got to work and it might sound strange, but networking gear isn't always reliable and what we've been able to achieve through our architectural approach and through our focus on automated testing has enabled us to produce a higher quality product which has been a major attractor of the enterprise customer. So you need to cover all those bases to succeed in this business. >> You're right, that network is absolutely essential. When anything goes down, whether it's a Facebook outage, it's world news. Tell me, what is the Arista advantage? >> The key advantage is the quality of our products. It's the fact that we have built an architecture that is more resilient to software and hardware errors. It's the way we test. We've made a tremendous investment in automated testing, so that our product has gone through hundreds of thousands of tests before it ever sees a customer. But actually the most important element behind quality, is the culture of your company, what do you believe? What's important to you? What gets you up in the morning? What are you thinking about and talking about to your employees? What's the most important thing, is it profitability? Is it making a deal, is it hitting a schedule? Or is it making sure the network works? We are 100% focused on that and it's been really gratifying to see the impact that's had. >> So last question, and thank you for speaking over the drum noise going on behind us, by the way, to get people into the auditorium. In terms of culture and the impact, what do you think this award means to your peers, your teams at Arista? >> Oh, it's just such an affirmation of the journey we've come through so far and the journey we still have ahead of us. We're very grateful for the award. >> So, I see so much momentum coming into 2019. What are some of the exciting things we can expect from Arista this year that you might be able to share with us? >> I think we're seeing a real transition from network designers focusing on the control plane of their network first to focusing on the management of the network first because management is actually the key to smooth operations. Our cloud vision product addresses that need. We're really excited about that transformation. >> Well, Kenneth, again, congratulations to Arista and yourself and your teams on the Outstanding Growth Award from ACG SV. We also thank you for spending some time with us on theCUBE. >> Thank you very much, it was my pleasure. >> I'm Lisa Martin and you're watching theCUBE. (energetic music)

>> Narrator: Live from Berlin, Germany, it's theCUBE. Covering NetApp Insight 2017. Brought to you by NetApp. (techno music) Welcome back to theCUBE's live coverage of NetApp Insight here in Berlin, Germany. I'm your host, Rebecca Knight, along with my cohost Peter Burris. We have two guests for this segment. We have Matt Watts, he is the director and data strategist and director of technology at NetApp, and Kenneth Cukier, a senior editor at The Economist, and author of the best-selling book Big Data, and author of a soon to be best-selling book on AI. Welcome. Thank you. Thank you much for coming on the show. Pleasure to be here. So, this is the, we keep hearing NetApp saying this is the day of the data visionary. I'd love to hear both of you talk about what a data visionary is, and why companies, why this is a necessary role in today's companies. Okay, so I think if you look at the generations that we've been through in the late nineties, early 2000's, it was all about infrastructure with a little bit of application and some data associated to it. And then as we kind of rolled forward to the next decade the infrastructure discussion became less. It became more about the applications and increasingly more about the data. And if we look at the current decade that we're in right now, the infrastructure discussions have become less, and less, and less. We're still talking about applications, but the focus is on data. And what we haven't seen so much of during that time is the roles changing. We still have a lot of infrastructure people doing infrastructure roles, a lot of application people doing application roles. But the real value in this explosion of data that we're seeing is in the data. And it's time now that companies really look to put data visionaries, people like that in place to understand how do we exploit it, how do we use it, what should we gather, what could we do with the information that we do gather. And so I think the timing is just right now for people to be really considering that. Yeah, I would build on what Matt just said. That, functionally in the business and the enterprise we have the user of data, and we have the professional who collected the data. And sometimes we had a statistician who would analyze it. But pass it along to the user who is an executive, who is an MBA, who is the person who thinks with data and is going to present it to the board or to make a decision based on it. But that person isn't a specialist on data. That person probably doesn't, maybe doesn't even know math. And the person is thinking about the broader issues related to the company. The strategic imperatives. Maybe he speaks some languages, maybe he's a very good salesperson. There's no one in the middle, at least up until now, who can actually play that role of taking the data from the level of the bits and the bytes and in the weeds and the level of the infrastructure, and teasing out the value, and then translating it into the business strategy that can actually move the company along. Now, sometimes those people are going to actually move up the hierarchy themselves and become the executive. But they need not. Right now, there's so much data that's untapped you can still have this function of a person who bridges the world of being in the weeds with the infrastructure and with the data itself, and the larger broader executives suite that need to actually use that data. We've never had that function before, but we need to have it now. So, let me test you guys. Test something in you guys. So what I like to say is, we're at the middle of a significant break in the history of computing. The first 50 years or so it was known process, unknown technology. And so we threw all our time and attention at understanding the technology. >> Matt: Yeah. We knew accounting, we knew HR, we even knew supply-chain, because case law allowed us to decide where a title was when. [Matt] Yep. But today, we're unknown process, known technology. It's going to look like the cloud. Now, the details are always got to be worked out, but increasingly we are, we don't know the process. And so we're on a road map of discovery that is provided by data. Do you guys agree with that? So I would agree, but I'd make a nuance which is I think that's a very nice way of conceptualizing, and I don't disagree. But I would actually say that at the frontier the technology is still unknown as well. The algorithms are changing, the use cases, which you're pointing out, the processes are still, are now unknown, and I think that's a really important way to think about it, because suddenly a lot of possibility opens up when you admit that the processes are unknown because it's not going to look like the way it looked in the past. But I think for most people the technology's unknown because the frontier is changing so quickly. What we're doing with image recognition and voice recognition today is so different than it was just three years ago. Deep learning and reinforcement learning. Well it's going to require armies of people to understand that. Well, tell me about it. This is the full-- Is it? For the most, yes it's a full employment act for data scientists today, and I don't see that changing for a generation. So, everyone says oh what are we going to teach our kids? Well teach them math, teach them stats, teach them some coding. There's going to be a huge need. All you have to do is look at the society. Look at the world and think about what share of it is actually done well, optimized for outcomes that we all agree with. I would say it's probably between, it's in single percents. Probably between 1% and 5% of the world is optimized. One small example: medical science. We collect a lot of data in medicine. Do we use it? No. It's the biggest scandal going on in the world. If patients and citizens really understood the degree to which medical science is still trial and error based on the gumption of the human mind of a doctor and a nurse rather than the data that they actually already collect but don't reuse. There would be Congressional hearings everyday. People, there would be revolutions in the street because, here it is the duty of care of medical practitioners is simply not being upheld. Yeah, I'd take exception to that. Just, not to spend too much time on this, but at the end of the day, the fundamental role of the doctor is to reduce the uncertainty and the fear and the consequences of the patient. >> Kenneth: By any means necessary and they are not doing that. Hold on. You're absolutely right that the process of diagnosing and the process of treatment from a technical standpoint would be better. But there's still the human aspect of actually taking care of somebody. Yeah, I think that's true, and think there is something of the hand of the healer, but I think we're practicing a form of medicine that looks closer to black magic than it does today to science. Bring me the data scientist. >> Peter: Alright. And I think an interesting kind of parallel to that is when you jump on a plane, how often do you think the pilot actually lands that plane? He doesn't. No. Thank you. So, you still need somebody there. Yeah. But still need somebody as the oversight, as that kind of to make a judgment on. So I'm going to unify your story, my father was a cardiologist who was also a flight surgeon in the Air Force in the U.S., and was one of the few people that was empowered by the airline pilots association to determine whether or not someone was fit to fly. >> Matt: Right. And so my dad used to say that he is more worried about the health of a bus driver than he is of an airline pilot. That's great. So, in other words we've been gah-zumped by someone who's father was both a doctor and a pilot. You can't do better than that. So it turns out that we do want Sully on the Hudson, when things go awry. But in most cases I think we need this blend of the data on one side and the human on the other. The idea that the data just because we're going to go in the world of artificial intelligence machine learning is going to mean jobs will be eradicated left and right. I think that's a simplification. I think that the nuance that's much more real is that we're going to live in a hybrid world in which we're going to have human beings using data in much more impressive ways than they've ever done it before. So, talk about that. I mean I think you have made this compelling case that we have this huge need for data and this explosion of data plus the human judgment that is needed to either diagnose an illness or whether or not someone is fit to fly a plane. So then where are we going in terms of this data visionary and in terms of say more of a need for AI? Yeah. Well if you take a look at medicine, what we would have is, the diagnosis would probably be done say for a pathology exam by the algorithm. But then, the health care coach, the doctor will intervene and will have to both interpret this for, first of what it means, translate it to the patient, and then discuss with the patient the trade-offs in terms of their lifestyle choices. For some people, surgery is the right answer. For others, you might not want to do that. And, it's always different with all of the patients in terms of their age, in terms of whether they have children or not, whether they want the potential of complications. It's never so obvious. Just as we do that, or we will do that in medicine, we're going to do that in business as well. Because we're going to take data that we never had about decisions should we go into this market or that market. Should we take a risk and gamble with this product a little bit further, even though we're not having a lot of sales because the profit margins are so good on it. There's no algorithm that can tell you that. And in fact you really want the intellectual ambition and the thirst for risk taking of the human being that defies the data with an instinct that I think it's the right thing to do. And even if we're going to have failures with that, and we will, we'll have out-performance. And that's what we want as well. Because society advances by individual passions, not by whatever the spreadsheet says. Okay. Well there is this issue of agency right? So at the end of the day a human being can get fired, a machine cannot. A machine, in the U.S. anyway, software is covered under the legal strictures of copywriting. Which means it's a speech act. So, what do you do in circumstances where you need to point a finger at something for making a stupid mistake. You keep coming back to the human being. So there is going to be an interesting interplay over the next few years of how this is going to play out. So how is this working, or what's the impact on NetApp as you work with your customers on this stuff? So I think you've got the AI, ML, that's kind of one kind of discussion. And that can lead you into all sorts of rat holes or other discussions around well how do we make decisions, how do we trust it to make decisions, there's a whole aspect that you have to discuss around that. I think if you just bring it back to businesses in general, all the businesses that we look at are looking at new ways of creating new opportunities, new business models, and they're all collecting data. I mean we know the story about General Electric. Used to sell jet engines and now it's much more about what can we do with the data that we collect from the jet engines. So that's finding a new business model. And then you vote with a human role in that as well, is well is there a business model there? We can gather all of this information. We can collect it, we can refine it, we can sort it, but is there actually a new business model there? And I think it's those kind of things that are inspiring us as a company to say well we could uncover something incredible here. If we could unlock that data, we could make sure it's where it needs to be when it needs to be there. You have the resources to bring to bed to be able to extract value from it, you might find a new business model. And I think that's the aspect that I think is of real interest to us going forward, and kind of inspires a lot of what we're doing. Great. Kenneth, Matt, thank you so much for coming on the show. It was a really fun conversation. Thank you. Thank you for having us. We will have more from NetApp Insight just after this. (techno music)

>> Welcome everybody, we're at the Flink Forward conference in San Francisco, at the Kabuki Hotel. Flink Forward U.S. is the first U.S. user conference for the Flink community sponsored by data Artisans, the creators of Flink, and we're here with special guest Kenneth Knowles-- >> Hi. >> Who works for Google and who heads up the Apache Beam Team where, just to set context, Beam is the API Or STK on which developers can build stream processing apps that can be supported by Google's Dataflow, Apache Flink, Spark, Apex, among other future products that'll come along. Ken, why don't you tell us, what was the genesis of Beam, and why did Google open up sort of the API to it. >> So, I can speak as an Apache Beam Team PMC member, that the genesis came from a combined code donation to Apache from Google Cloud Dataflow STK and there was also already written by data Artisans a Flink runner for that, which already included some portability hooks, and then there was also a runner for Spark that was written by some folks at PayPal. And so, sort of those three efforts pointed out that it was a good time to have a unified model for these DAG-based computational... I guess it's a DAG-based computational model. >> Okay, so I want to pause you for a moment. >> Yeah. >> And generally, we try to avoid being rude and cutting off our guests but, in this case, help us understand what a DAG is, and why it's so important. >> Okay, so a DAG is a directed acyclic graph, and, in some sense, if you draw a boxes and arrows diagram of your computation where you say "I read some data from here," and it goes through some filters and then I do a join and then I write it somewhere. These all end up looking what they call the DAG just because of the fact that it is the structure, and all computation sort of can be modeled this way, and in particular, these massively parallel computations profit a lot from being modeled this way as opposed to MapReduce because the fact that you have access to the entire DAG means you can perform transformations and optimizations and you have more opportunities for executing it in different ways. >> Oh, in other words, because you can see the big picture you can find, like, the shortest path as opposed to I've got to do this step, I've got to do this step and this step. >> Yeah, it's exactly like that, you're not constrained to sort of, the person writing the program knows what it is that they want to compute, and then, you know, you have very smart people writing the optimizer and the execution engine. So it may execute an entirely different way, so for example, if you're doing a summation, right, rather than shuffling all your data to one place and summing there, maybe you do some partial summations, and then you just shuffle accumulators to one place, and finish the summation, right? >> Okay, now let me bump you up a couple levels >> Yeah. >> And tell us, so, MapReduce was a trees within the forest approach, you know, lots of seeing just what's a couple feet ahead of you. And now we have the big picture that allows you to find the best path, perhaps, one way of saying it. Tell us though, with Google or with others who are using Beam-compatible applications, what new class of solutions can they build that you wouldn't have done with MapReduce before? >> Well, I guess there's... There's two main aspects to Beam that I would emphasize, there's the portability, so you can write this application without having to commit to which backend you're going to run it on. And there's... There's also the unification of streaming and batch which is not present in a number of backends, and Beam as this layer sort of makes it very easy to use sort of batch-style computation and streaming-style computation in the same pipeline. And actually I said there was two things, the third thing that actually really opens things up is that Beam is not just a portability layer across backends, it's also a portability layer across languages, so, something that really only has preliminary support on a lot of systems is Python, so, for example, Beam has a Python STK where you write a DAG description of your computation in Python, and via Beam's portability API's, one of these sort of usually Java-centric engines would be able to run that Python pipeline. >> Okay, so-- >> So, did I answer your question? >> Yes, yes, but let's go one level deeper, which is, if MapReduce, if its sweet spot was web crawl indexing in batch mode, what are some of the things that are now possible with a Beam-style platform that supports Beam, you know, underneath it, that can do this direct acyclic graph processing? >> I guess what I, I'm still learning all the different things that you can do with this style of computation, and the truth is it's just extremely general, right? You can set up a DAG, and there's a lot of talks here at Flink Forward about using a stream processor to do high frequency trading or fraud detection. And those are completely different even though they're in the same model of computation as, you know, you would still use it for things like crawling the web and doing PageRank over. Actually, at the moment we don't have iterative computations so we wouldn't do PageRank today. >> So, is it considered a complete replacement, and then new used cases for older style frameworks like MapReduce, or is it a complement for things where you want to do more with data in motion or lower latency? >> It is absolutely intended as a full replacement for MapReduce, yes, like, if you're thinking about writing a MapReduce pipeline, instead you should write a Beam pipeline, and then you should benchmark it on different Beam backends, right? >> And, so, working with Spark, working with Flink, how are they, in terms of implementing the full richness of the Beam-interface relative to the Google product Dataflow, from which I assumed Beam was derived? >> So, all of the different backends exist in sort of different states as far as implementing the full model. One thing I really want to emphasize is that Beam is not trying to take the intersection on all of these, right? And I think that your question already shows that you know this, we keep sort of a matrix on our website where we say, "Okay there's all these different "features you might want, "and then there's all these backends "you might want to run it on," and it's sort of there's can you do it, can you do it sometimes, and notes about that, we want this whole matrix to be, yes, you can use all of the model on Flink, all of it on Spark, all of it on Google Cloud Dataflow, but so they all have some gaps and I guess, yeah, we're really welcoming contributors in that space. >> So, for someone whose been around for a long time, you might think of it as an ODBC driver, where the capabilities of the databases behind it are different, and so the drivers can only support some subset of a full capability. >> Yeah, I think that there's, so, I'm not familiar enough with ODBC to say absolutely yes, absolutely no, but yes, it's that sort of a thing, it's like the JVM has many languages on it and ODBC provides this generic database abstraction. >> Is Google's goal with Beam API to make it so that customers demand a level of portability that goes not just for the on-prim products but for products that are in other public clouds, and sort of pry open the API lock in? >> So, I can't say what Google's goals are, but I can certainly say that Beam's goals are that nobody's going to be locked into a particular backend. >> Okay. >> I mean, I can't even say what Beam's goals are, sorry, those are my goals, I can speak for myself. >> Is Beam seeing so far adoption by the sort of big consumer internet companies, or has it started to spread to mainstream enterprises, or is still a little immature? >> I think Beam's still a little bit less mature than that, we're heading into our first stable release, so, we began incubating it as an Apache project about a year ago, and then, around the beginning of the new year, actually right at the end of 2016, we graduated to be an Apache top level project, so right now we're sort of on the road from we've become a top level project, we're seeing contributions ramp up dramatically, and we're aiming for a stable release as soon as possible, our next release we expect to be a stable API that we would encourage users and enterprises to adopt I think. >> Okay, and that's when we would see it in production form on the Google Cloud platform? >> Well, so the thing is that the code and the backends behind it are all very mature, but, right now, we're still sort of like, I don't know how to say it, we're polishing the edges, right, it's still got a lot of rough edges and you might encounter them if you're trying it out right now and things might change out from under you before we make our stable release. >> Understood. >> Yep. All right. Kenneth, thank you for joining us, and for the update on the Beam project and we'll be looking for that and seeing its progress over the next few months. >> Great. Thanks for having me. >> With that, I'm George Gilbert, I'm with Kenneth Knowles, we're at the dataArtisan's Flink Forward user conference in San Francisco at the Kabuki Hotel and we'll be back after a few minutes.

live from Barcelona Spain it's the cube covering Cisco live 2020s brought to you by Cisco and its ecosystem partners welcome back to Barcelona Spain everybody this is the cube the leader and live tech coverage and we're here day one for us at Cisco live Barcelona even though we did a little preview game preview yesterday my good friend kena Reilly is here he's the director of customer experience at Cisco and he's joined by Kyle winters Technical Marketing engineer for the customer experience technology and transformation group it's six to go guys great to see you thanks for coming on and you know we love talking customer experience Cisco is a it's a big company big portfolio and a lot of complexity for clients and so bring it all together and customer experience is very important can't we have it a conversation with Alastair early today and he was talking about Cisco's commitment from the top chuck Robbins on down to really improve that customer experience bring essentially a digital virtual experience to your customers and you guys obviously fit into that right absolutely so about two years ago when Chuck brought in Maria Martinez that was the first step into really pushing Cisco to focus more on successful outcomes for customers so we had already always sold that way but with the complexity of technology and how fast technology is moving accelerating value realization for customers has never been bigger especially in the security space because as we've talked before you know with everything that goes on today and the fact that the bad guys are trying to get data faster quicker and different getting the technology in play operational and production it has never been more important and we're gonna dig in with Kyle with some detail and double click into the lifecycle specifically and the different points of that journey but that's really important for any customer experience is really understanding that lifecycle that maturity model can you talk about that a little bit yeah so so with us you know we've been at it for about six years when we started as Lancope so we've got a great model and you know our approach to getting outcomes for customers is completely in line with with the strategy of our products and technologies and all security so it's really important that you align with that strategy because salespeople sell and they sell you the what we sell the how we're gonna get you and so you have to understand what it is that customers need and how that technology maps because you don't want a shelf where and you don't want products or technology sitting there waiting to be implemented because you know these days especially with the move to the cloud it's got to get up and running you know within an hour so our model has always been that way we built our model with customer first and so we are you know we are the security experts we're the trusted security adviser so when we go in and work with customers we completely know exactly those outcomes that they need and with all the sort of technologies and products that we have not only with stealthWatch but the other products that sent ulema tree to us we have in Kyle will talk about how our service is completely aligned with those outcomes and the journeys that we will take our customers on yes a faster adoption means faster time to value obviously let's focus in on stealthWatch Kenneth you came in with the stealthWatch acquisitions been very successful I mean Cisco security business grew 22% last quarter we'll talk more about the sort of umbrella but let's drill in with Kyle to stealthWatch services specifically maybe you could sort of take us through you know at a high level what what the areas are and then we can sort of follow up on yeah yes so so our customer maturity model when it comes to services there's kind of three different stages to it it starts with the visibility stage so we have services around being able to deploy an operational I stealthWatch will bring in our best practices and help customers get up to speed and using the system quickly and efficiently from there we also have services around detection capabilities so being able to use automation and integrations to further the detection capabilities of stealthWatch things like being able to classify host groups through automation from source like IP address management systems things like asset discovering classification service that helped drive segmentation efforts all of these things help improve the behavioral algorithms and processes that stealthWatch is using to detect these threats in real time and then from there we have an integration stage as well - which is all about bridging the gap between stealthWatch and the rest of not only Cisco's portfolio but the entirety of our customer security portfolio as well and some of those services include things like sim integrations being able to integrate stealthWatch with Splunk we have services such as our proxy integration service as well a lot of different types of services that we're able to help get our customers to the next stage with their stealth watch environments I got a lot of questions yeah we could get to it and you guys could take it by stage so yes the sort of visibility that's where you start that's when you do the discovery right so what what are you discovering how do you actually do that discovery so a lot of that is about making sure that we've got all the flow and telemetry that we need from the various different sources of our network coming into stealthWatch feeding into the processes and algorithms that are going on there so a lot of things is not only net flow data but getting ice integrated in there as well being able to pull that user attribution data and being able to find sources of data where we maybe can convert it into net flow if it's not already net flow and be able to ingest that data as well we also in that space typically to help set up customers with a lot of different best practices that kind of get them operationalized very quickly and things like being able to build custom reports and dashboards for them will work through them which is kind of understanding the system from a base level to more of a professional fully operational level a lot of times we come in during the stage two and customers don't even understand what's going on in their network they're seeing things that maybe they've never seen before one stealthWatch turns on a great example actually as we were at a large financial firm and we were able within 30 minutes of being on site with them through our services team we were able to identify rogue DNS servers unsecured telnet going on sequel injections suspicious SMB and that's the sage traffic this is all just within 30 minutes of us coming on there and taking a look at this stuff you don't even want to look at sometimes yeah so who's doing this can I mean is this sort of all automated you've got professionals sort of overseeing it in our society yeah so the team that we have the technology transformation team when we've talked about it before that team is kind of on the bleeding edge of helping customers and you know a lot of these services that that Kyle talked about is we are building services that customers are consuming based on their needs today and that's why the team is very flexible we build you know a lot of these integrations with those requirements in mind and then we take those and we can scale that so these are all field engineers we have developers so in in essence it is like a mini development team that goes out and works on the specific things that customers need to protect themselves okay and my understanding is there's a there's an ongoing learning with the customers and a it's a transfer of knowledge from day one right there the customer is with you on this in each of these phases and you're sort of learning as they go along and that's sort of part of the transfer of knowledge it's I would say even a tool a transfer knowledge too because we're teaching them our best practices and how to best be successful with these systems but we also learn from them what's going on what are the trends that they're seeing how can we help get them to the next stage and that's where our technology and transformation group comes and they're able to be on the cutting edge here the problems that the customers are talking about and be able to take stealthWatch to the next level okay let's dig it to the detection phase so this is where you're classifying things like host groups etc I'm interested in how that happens is that you know it used to be you'd get everybody in a room you start drawing pictures and that just doesn't scale it's too complicated today so can you auto classify stuff how does that all work and use them oh yeah genius math to do that so so traditionally the the you know the MIT's a manual effort to classify your whole group somebody who's very familiar with the network comes in and they say okay these are the DNS servers these are the web servers these are this network scanners oh oh today but the problem is that today's networks are so dynamic and fluid that what the network looks like today is not necessarily going to be the same tomorrow so there needs to be that relief from the analyst to be able to come in there needs to be that automation that they can go in each day and know that their system is going to be classified accurately and meaningfully that way the behavioral detection that is built into stealthWatch is also driven and accurate and meaningful - so we have this service so for example our host group automation service and through that we're able to pull in telemetry and data from various different sources such as IP address management systems cmdbs we can do threat feeds as well external threat feeds and we're able to drive the classification based off of the metadata that we see from these different sources so we're able to write different types of automation rules that essentially pull this data in detect the different patterns that we're seeing with that metadata and then drive that classification stealthWatch that way when you come in that next day you know that your network scanners are gonna be classified as Network scanners and your web servers are gonna be web servers etc etc so you you have that integrity of data coming in every single day yeah so a lot of different data sources data quality obviously really important I mean you'd love it if somebody had like you know a single CMDB from ServiceNow boom and pop it right in but that's not always the case we never always the case there's always a challenge and that's where kind of our services engineers come in they're able to work through these different environments and understand what the main admit what the metadata is where we need to go and how we need to classify and driving the classification from there so it does require a little bit of a human element on the front-end but once we get it worked out it can be fully automated you know there's lots of different sources and the quality of the data is not always there we've seen for example customers who have Excel spreadsheets and everything is just you're all over the place and we have to figure out a way to work with that and that's part of what our engineer success is so before we get to the integration piece can you been following this industry for for a while um security is really exciting space it's growing like crazy it's really hard I did a braking analysis piece you know a few weeks ago just talking about the fragmentation in the business you see startups coming out like crazy big valuations at the same time you see companies like Cisco with big portfolios yeah you mentioned Splunk before and they've kind of become a gold standard for for log files but very complex and you talk to security practitioners and they'll tell you our number one problem is just skillsets so get you know paint a picture of what's going on in the security world and what's in the house cisco is trying to address that so the security teams the analysts all the way up the management chain to the sea so they're under tremendous pressure their businesses are growing and so when their businesses are growing the sort of a tax base is growing and the business is growing faster than they can protect it so with the sort of increase in the economy more money more investment to build more point products so you've got a very stressed team a lot of turnover skill sets aren't great and what do we do as an industry we just give them more technology right more tools more tools complexity avalanche ok they're buried all right so we feel and we've made great strides within the security group within Cisco is we're taking the products that we have and we're integrating them under one platform so that it is in a bunch of point products and so that the that's what everybody else is doing I mean the other guys are acquiring companies then they're trying to integrate those because the customers are saying I don't need another point protocol yeah yeah it's too much so you know with us that's the way we approach it and now with the platform that's going to be launching this year the cisco threat response that we've launched you're gonna see later on in this year that we will be selling and positioned in implementing the entire platform yeah so I have a stat I came up with this and my one of my analyses it was the the worldwide economy is like 86 trillion and we spent about 0.014 percent on security so we're barely scratching the surface so this sort of tools avalanche probably isn't gonna change though integration becomes an extremely important aspect of the customer journeys and it's through that and to continue on that point you just made as well - I believe in our Cisco cybersecurity report from 2017 only fifty four six percent or fifty seven percent of actual threats are being investigated remediated so there's always that need to kind of help build bridge that gap make it easier for people to understand these threats and and mitigate and prioritize know what to go after right which part the integration exactly so we do have a lot of different integration services as well - for example I mentioned our sim integration service one thing that we can really do that's really awesome with that is we're able to deploy for example with Splunk a full-fledged stealthWatch for Splunk application that allows you to utilize stealth watches capabilities directly inside of Splunk without having to actually store an index any data inside of Splunk so all these api's are on demand inside of this app and available throughout the rest of the Splunk capabilities as well so you can extend it into other search reporting correlate that against other sets of data that you have and Splunk you can do quite a bit with it we also have other ways absolutely advantage of that is just obviously integration you're not leaving the environment plus its cost you're saving customers money a lot of a lot of customers kind of see their sim as a single pane of glass so being able to bring that stealthWatch value into that single pane is a huge win for our customers not to mention that reduction in licensing costs as well we have other ways to that we can reduce licensing costs some customers like to send their flow data into their sim for deeper analytics and long-term retention and we have a service we call it our flow adapter service and through this service we're essentially able to take buy flow off of the stealthWatch flow collectors and the buy flow is essentially when the raw net flow hits the stealthWatch flow collectors it's coming from multiple different routers and switches on the network this is gets converted into bi flow which is bi-directional deduplicated stitched together flow records so right there by sending that data into a sim or a data Lake as opposed to ronette flow we see data reduction cost anywhere from 15 to 80% depending on how the customers network is architected great any any favorite customer examples you have that you can share where ya guys have gone in you know provided these services and and it's had an outcome that got the customer excited or you found some bad guys or there's one that's one of my favorites so we have this service we call it our asset discovering classification service and I mentioned the host tree of automation service that's if you have some sort of authoritative source we can pull that information in but if a customer doesn't have that authoritative source they don't know what's on their network and a lot of times too they want to do a segmentation effort they're undergoing network segmentation but they need to understand what's on their network how these devices are communicating and that's where our asset discovery classification service comes in we're able to pull in telemetry not just from stealthWatch but other sources such as ice tetration Active Directory I Pam's again as well and we're able to essentially profile these different devices based off of the nature of their behavior so we were at a kind of a large technology company and we were essentially in this effort trying to segment their security cameras and upon segmenting their security cameras we were able to build this report where we can see the security camera and how its communicating with the other parts of the network and we noticed that there was essentially two IP addresses from inside of their network that were accessing all these different security cameras but they were not authorized to so with this service we were able to see that these different these two hosts were unauthorized actually accessing these devices that got reported up through the management chain and ultimately those two employees were no longer at that technology permanence that was discovered nice to love it alright bring us on we're here in the dev net zone sort of all about hit for structures code and software and and and and talk a little bit about the futures where you see this all going yeah so for us for Cisco security the future is really bright we've either built or acquired a portfolio that the customers really need that get absolute outcomes that customers need and through the customer experience organization certainly stealthWatch is fitting into the broader play to to get customers who have all those technologies get that operational and get them success so when we talked last summer I told you the jury was still out we would see how the journeys gonna go and the journey has started it has gotten much better since the summer and this year I think we're gonna be doing some great things for our customers just we can't get in too much of the business but stealthWatch customers are still expanding because I think we told you last time customers can never get enough stealthWatch okay the attack surface is too big right so so we we feel really good about that and the other technologies that they're building really fit into what customers need we're going to the cloud so they're gonna be able to consume cloud on-prem hybrid protect networks the campus protect their cloud infrastructure so we're really checking a lot of boxes in our group brings it all together and takes all the complexity out of that for customers just to get them the outcomes that I named us Cisco is one of my four star security companies for 2020 okay based on spending data that we share from our friends at ETR and the reason was because cisco has both a large presence in the market and but also you have spending momentum I mentioned 22% you know growth last quarter and the security business but you've also got the expertise you put your money where your mouth is you know the big portfolio which helps if you can bring it together and do these types of integrations it simplifies the customers environment and so that's a winner in my book so I named you along with some other high fliers right you know and you see some really interesting startups coming out and probably acquisition targets probably something that aren't your radar but guys thanks so much for coming on the cube thank you thank you I keep it right there everybody we'll be back with our next guest is a Dave Volante for the cubes 2 min Amanda John Faria are also in the house at Cisco live Barcelona right back

(upbeat electronic music) >> Announcer: Live, from Berlin, Germany it's theCUBE. Covering NetApp Insight 2017. Brought to you by NetApp. Welcome back to theCUBE's live coverage of NetApp Insight here in Berlin, Germany. I'm your host, Rebecca Knight, along with my co-host Peter Burris. We are joined by Deepak Visweswaraiah. He is the senior vice president for data fabric manageability at NetApp. Thanks so much for coming on the show, Deepak. Thank you. So let's talk about the data fabric, and why modern IT needs it to do what it needs to do. For acceleration. I think anyone attending the conference, I thought the keynote that happened yesterday Kenneth Corky from Economist actually talked about how data actually is growing. And then how much of that is becoming more and more important to companies. Not only just from an ability to be able to actually handle data, but how they make their decisions based on the amount of data that they have today. The fact that we have that technology, and we have the mindset to be able to actually handle that data, I think gives that unique power to customers who actually have that data. And within their capacity. So, if you look at it in terms of the amount of data growing and what companies are trying to do with that, the fact is that data is not all in one place, it's not all in one format, it's not all just sitting in some place. Right, in terms of the fact that we call it, you know, data being diverse, data being dynamic and then what have you. So, this data, for any CIO, if you talk to an IT organization and ask them in terms of do you even really know where all your data lives, they probably, you know, 80% of the time they don't know where it is all. And they do not know who is accessing what data. Do they actually really have the access or the right people accessing the right data? And then what have you. So, being able to look at all of this data in different silos that is there, to be able to have visibility across these, to be able to actually handle the diversity of that data, whether it is structured, unstructured, comes from, you know, the edges of the network, whether it is streaming, and different types of, you know, media for that matter, whether it is streaming, video, audios, what have you. With that kind of diversity in the data, and the fact that it lives in multiple places, how do you handle all of that in a seamless fashion? Having a ability to view all of that and making decisions on leveraging the value of that data. So, number one, is really to be able to handle that diversity. What you need is a data fabric that can actually see multiple end points and kind of bring that together in one way and one form with one view for a customer. That's the number one thing, if you will. The second thing is in terms of being able to take this data and do something that's valuable in terms of their decision making. How do I decide to do something with it? I think one of the examples you might have seen today for example, is that, we have 36 billion data points coming from our own customer base, that we bring back to NetApp, and help our customers to understand in the universe of the storage end points with all the data collected, we can actually tell them what may proactively tell them, what maybe going wrong what can actually they do better. And then how can they do this. This is really what that decision making capability is to be able to analyze. It's about being able to provide that data, for analytics to happen. And that analytics may happen whether it happens in the cloud, whether it happens where the data is, it shouldn't really matter, and it's our responsibility to provide or serve that data in the most optimized way to the applications that are analyzing that data. And that analysis actually helps make significant amount of decisions that the customers are actually looking to. The third is, with all of this that is underlying infrastructure that provides the capability to handle this large amount of data, not only, and also that diversity that I talked about. How do you provide that capability for our customers, to be able to go from today's infrastructure in their data center, to be able to have and handle a hybrid way of doing things in terms of their infrastructure that they use within their data center, whether they might actually have infrastructure in the cloud, and leveraging the cloud economics to be able to do what they do best, and, or have service providers and call locators, in terms of having infrastructure that may be. Ability to be able to seamlessly look all of that providing that technology to be able to modernize their data center or in the cloud seamlessly. To be able to handle that with our technology is really the primary purpose of data fabric. And then that's what I believe we provide to our customers. So, people talk about data as an asset. And folks talk about what you need to ensure the data becomes an asset. When we talk about materials we talk about inventory we talk about supply chain, which says there's a linear progression, one of the things that I find fascinating about the term fabric even though there's a technical connotation to it, is it does suggest that in fact what businesses need to do is literally weave a data tapestry that supports what the business is going to do. Because you cannot tell with any certainty it's certainly not a linear progression, but data is going to be connected in a lot of different ways >> Deepak: Yeah To achieve the goals of the business. Tell us a little bit about the processes the underlying technologies and how that informs the way businesses are starting to think about how data does connect? >> Deepak: Can you repeat the last part? How data connects, how businesses are connecting data from multiple sources? And turning it into a real tapestry for the business. Yeah, so as you said, data comes in from various different sources for that matter, in terms of we use mobile devices so much more in the modern era, you actually have data coming in from these kind of sources, or for example in terms of let's say IoT, in terms of sensors, that are all over the place in terms of how that data actually comes along. Now, let's say, in terms of if there is a customer or if there is an organization that is looking at this kind of data that is coming from multiple different sources all coming in to play the one thing is just the sheer magnitude of the data. What typically we have seen is that there is infrastructure at the edge, even if you take the example of internet of things. You try and process the data at the edge as much as you can, and bring back only what is aggregated and what is required back to you know, your data center or a cloud infrastructure or what have you. At the same time, just that data is not good enough because you have to connect that data with the internal data that you have about-- Okay, who is this data coming from and what kind of data, what is that meta-data that connects my customers to the data that is coming in? I can give you a couple of examples in terms of let's say there is an organization that provides weather data to farmers in the corners of a country that is densely populated, but you really can never get into with a data center infrastructure to those kind of remote areas. There are at the edge, where you have these sensors in terms of being able to sample the weather data. And sample also the data of the ground in itself, it terms of being able to, the ultimate goals is to be able to help the farmer in terms of when is the right time to be able to water his field. When is the right time to be able to sow the seeds. When is the right time for him to really cut the crops, when is the most optimized time. So, when this data actually comes back from each of these locations, it's all about being able to understand where this data is coming from, from the location, and being able to connect that to the weather data that is actually coming from the satellites and relating that and collating that to be able to determine and tell a farmer on his mobile device, to be able to say okay, here is the right time, and if you don't actually cut the crops in the next week, you may actually lose the window because of the weather patterns that they see and what have you. That's an example of what I could talk about as far as how do you connect that data that is coming in from various sources. And as a great example, I think, was at the keynote yesterday about a Stanford professor talking about the race track, it's really about that race track and not just about any race track that where the cars are actually making those laps, to be able to understand and predict correctly in terms of when to make that pit stop in a race. You really need the data from that particular race track because it has characteristics that have an impact on the wear and tear of the tires. For example. That's really all about being able to correlate that data. So it's having the understanding of the greater context but the specific context too. >> Deepak: Absolutely, absolutely. Great. You also talked about you talked about the technology that's necessary, but you also mentioned the right mindset. Can you unpack that a little bit for our viewers? The mindset I talked about earlier, was really more in terms of can we actually if you think some time before, we couldn't have attacked some of the problems that we can afford to today. It's really having the mindset of being able to from the data I can do things that I could never do before. We could solve, we can solve things in the nature of being able to being able to impact lives if you will. One of our customers leads a Mercy technology. Has built a out care platform, that provides that has a number of healthcare providers coming together. Where they were actually able to make a significant impact where they could actually determine 40% of the patients coming into their facilities, really were prevented from coming back into with a sepsis kind of diagnosis. Before then, they reduce that sepsis happening in 40% of the time. Which is a significant, significant impact, if you will, for the human. Just having that mindset in terms of you have all the data and you can actually change the world with that data, and you can actually find solutions to problems that you could never have before because you have the technology and you have that data. Which was never there before. So you can actually make those kinds of improvements. It's all about extracting those insights. >> Deepak: Absolutely. Thank you so much for coming on the show, Deepak. It was a pleasure having you Thank you for having me. Thank you very much. I'm Rebecca Knight, for Peter Burris, we will have more from NetApp Insight in just a little bit. (dramatic electronic music)

>> Announcer: It's The Cube. Covering HPE Big Data Conference 2016. Now here are your hosts, Dave Vellante and Paul Gillin. >> Welcome back to Boston, everybody, this is The Cube, we're here live at HP's big data conference, hashtag seize the data. Steve Spear is here, he's an author, MIT professor, author of The High Velocity Edge, welcome to The Cube, thanks for coming on. >> Oh, thanks for having me. >> I got to tell you, following Phil Black, you were coming onstage, I have never heard you speak before, I said, "Oh, this poor guy," and you did awesome, you were great, you held the audience, so congratulations, you were very dynamic and he was unbelievable and you were fantastic, so. >> Today was second-worst speaking setup, one time I was on a panel where it was three admirals, a general, and then the other guy wearing a suit, I said, "Well at least another schmo in a suit," and his opening lines were, "You know, this reminds me, "when I was on the space shuttle and we were flying "to the Hubble," and I'm like, "A flipping astronaut, "I got to follow an astronaut?" So anyway, this was only a SEAL, there were a lot of them, there were far fewer astronauts, so that was easy. >> What I really liked about your talk is, first of all, you told the story of Toyota, which I didn't know, you may. >> No, my experience with Toyota was in the early '70s, I remember the Toyota sort of sweeping into the market but you talked about 20 years before it when they were first entering and how this really was a company that had a lot of quality problems and it was perceived as not being very competitive. >> Yeah, Toyota now people look at as almost, they just take for granted the quality, the productivity, they assume good labor relations and that kind of thing, it's non-unionized, not because the unions haven't tried to unionize, but the employees don't feel the need. And again, in the '50s, Toyota was absolutely an abysmal auto-maker, their product was terrible, their productivity was awful and they didn't have particularly good relations with the workforce either. I mean, it's a profound transformation. >> And you gave this test, in the 50s, I forget what it was, it was one-tenth the productivity of the sort of average automobile manufacturer and then they reached parity in '62, by '68 they were 2X, and by '73, they were off the charts. >> Right, right, right. >> Right, so amazing transformation and then you try to figure out how they did it and they couldn't answer, but they said, "We can show you," right? And that sort of led to your research and your book. >> Yeah, so the quick background is in some regards, this fellow Kenneth Bowen, who was my mentor and advisor when I was doing my doctorate, he could argue we were late to the game because people started recognizing Toyota as this paragon of virtue, high quality at low cost, and so that in the 1980s prompted this whole investigation and the term lean manufacturing came out of the realization that on any given day, Toyota and suppliers were making basically twice the product with half the effort and so you had this period of '85 to about '95 where there was this intense attempt to study Toyota, document Toyota, imitate Toyota, General Motors had a joint venture with Toyota, and then you have the mid-'90s and there's no second Toyota, despite all this investment, so we go to the Toyota guys and say, "Look, clearly if everyone is studying you, imitating you, "copying you, and they haven't replicated you, "they've missed something, so what is it?" And they say, "I'm sorry, but we can't tell you." And we said, "Well you got to be kidding, I mean, "you have a joint venture with your biggest competitor, "General Motors," and they said, "No, no, it's not that we wouldn't tell you, "we just actually don't know how to explain what we do "'cause most of us learn it in this very immersive setting, "but if you'd like to learn it, "you can learn it the way we do." I didn't realize at the time that it would be this Karate Kid wax-on, wax-off, paint-up, paint-down experience, which took years and years to learn and there are some funny anecdotes about it but even at the end, their inability to say what it is, so I went years trying to capture what they were doing and realizing I was wrong 'cause different things wouldn't work quite right, and I can tell you, I was on the Shinkansen with the guy who was my Toyota mentor and I finally said, "Mr. Oba, I think I finally "figured it out, it all boils down to these basic "approaches to seeing and solving problems." And he's looking over my cartoons and stuff and he says, "Well, I don't see anything wrong with this." (laughs) >> That was as good as it got. >> That was as good as it got, I was like, "Score, nothing wrong that he can see!" So anyway. >> But so if you talk about productivity, reliability, you made huge gains there, and the speed of product cycles, were the three knobs that Toyota was turning much more significantly than anybody else and then fuel efficiency came. >> Right, so if you start looking at Toyota and I think this is where people first got the attraction and then sort of the dismissive of, we don't make cars, so the initial hook was the affordable reliability, they could deliver a much higher-quality car, much more affordable based on their productivity. And so that's what triggered attention which then manifest itself as this lean manufacturing and its production control tools. What then sort of started to fall off people's radar is that Toyota not only stayed ahead on those dimensions but they added to the dimensionality of the game, so they started introducing new product faster than anybody else and then they introduced new brand more successfully so all the Japanese, Nissan, Honda, Toyota, all came out with a luxury version, but no one came out with Lexus other than Toyota. The Affinity and the Acura, I mean, it's nice cars, but it didn't become this dominant brand like the Lexus. And then in trying to hit the youth market, everyone tried to come up with, like Honda had the Element but nothing like the Scion, so then Toyota's, and that's much further upstream, a much more big an undertaking than just productivity in a factory. And then when it came time to this issue around fuel efficiency, that's a big technology play of trying to figure out how you get these hybridized technologies with a very very complex software engineering overlay to coordinate power flow in this thing and that, and everyone has their version of hybrid, but no one has it through six generations, 21 platforms, and millions of copies sold. So it didn't matter where you were, Toyota figured out how to compete on this value to market with speed and ease which no one else in their industry was replicating. >> You're talking about, this has nothing to do with operational efficiency, when you talk about the Scion for example, you're talking about tapping into a customer, into an emotional connection with your customer and being able to actually anticipate what they will want before they even know, how do you operationalize that? >> So I think, again, Toyota made such an impression on people with operational efficiency that a lot of their genius went unrecognized, so what I was trying to elaborate on this morning is that Toyota's operational efficiency is not the consequence of just more clever design of operations, like you have an algorithm which I lack and so you get to a better answer than I do, it was this very intense almost empathetic approach to improving existing operations, so you're working on something and it's difficult so we're perceptive of that difficulty and try to understand the source of that difficulty and resolve it, and just do that relentlessly about everything all the time, and it's that empathy to understand your difficulty which then becomes the trigger for making things better, so as far as the Scion comes in, what you see is the same notion of empathic design apply to the needs of the youth market. And the youth market unlike the folks who are, let's say at the time, middle-aged, was less about reliable affordability, but these were people who were coming of age during the Bannatyne era where, very fast mass customization or the iPod era, which was common Chassis but very fast, inexpensive personalization and the folks at Toyota said, "You know what, "the youth market, we don't really understand that, "we've been really successful for this older mid-market, "so let's try to understand the problems that the youth "are trying to solve with their acquisitions," and it turned out personalization. And so if you look at the Scion, it wasn't necessarily a technically or technologically sophisticated quote-unquote sexy product, what it did was it leant itself towards very diverse personalization, which was the problem that the youth market was trying to solve. And you actually see, if I can go on this notion of empathic design, so you see this with the Lexus, so I think the conventional wisdom about luxury cars was Uber technology and bling it, throw chrome and leather and wood and when Toyota tried that initially, they took what was I guess now the Avalon, full-sized car, and they blinged it up and it was contradictory 'cause if you're looking for a luxury car, you don't go to a Toyota dealer, and if you go to a Toyota dealer and you see something with chrome and leather and wood veneer, you're like, you have dissonance. So they tried to understand what luxury meant from the American consumer perspective and again, it wasn't, you always wish you'd get this job, but they sent an engineering team to live in Beverly Hills for some months. (laughs) It's like, ooh, twist my arm on that one, right? But what they found was that luxury wasn't just the physical product, it was the respectful service around it, like when you came back to your hotel room, you walked in, people remembered your name or remembered that, oh we noticed that you used a lot of bath towels so we made sure there were extra in your room, that sort of thing, and if you look at the Lexus, and people were dismissive of the Lexus, saying, "It looks like slightly fancier Toyota, "but what's the big deal, it's not a Beamer or Mercedes." But that wasn't the point, it was the experience you got when you went for sales and service, which was, you got treated so nice, and again, not like hoity toity but you got treated respectfully, so anyway, it all comes back to this empathic design around what problem is the customer or someone inside a plan trying to solve. >> So Toyota and Volkswagen trying to vie for top market share but Toyota, as you say, has got this brand and this empathy that Volkswagen doesn't. You must get a lot of questions about Tesla. Thoughts on Tesla. >> Yeah, cool product, cool technology and time will tell if they're actually solving a real problem. And I don't mean to be dismissive, it's just not an area where I've spent a lot of time. >> And we don't really know, I mean, it's amazing and a software-defined automobile and autonomous, very difficult to predict, we're very tight on time. >> All the cool people seem to drive them though. >> Yeah, that's true. Last question I have is, what the heck does this have to do with analytics at a conference like this? >> Right, so you start thinking about the Toyota model, really, it's not that you can sit down and design something right, it's that you design things which you know deep-rooted in your DNA is that what you've designed is wrong, and that in order to get it right and actually much righter than anything else in the marketplace, what you need to do is understand what's wrong about it and so the experience of the user will help inform what's wrong, the worker rounds they do, the inconveniences they experience, the coping, the compensation they do, and that you can not only use that to help inform what's wrong, but then help shape your understanding of how to get to right, and so where all this fits in is that when you start thinking about data, well first of all, these are gigantic systems, right, which it's probably well-informed to think in terms of these systems are being designed by flawed human beings so the systems themselves have flaws, so it's good to be attentive to the flaws that are designed in it so you can fix them and make them more usable by your intended clientele. But the other thing is that these systems can help you gain much greater precision, granularity, frequency of sampling and understanding of where things are misfiring sooner than later, smaller than larger, so you can adjust and adapt and be more agile in shaping the experience. >> Well Steve, great work, thanks very much for coming on The Cube and sharing and great to meet you. >> Yeah likewise, thanks for having me. >> You're welcome. Alright, keep it right there, everybody, Paul and I will be back with our next guest, we're live from Boston, this is The Cube, we'll be right back. (upbeat music)

>>Hi buddy, we're back. This is Dave Volante with the cube goes out to the shows. We extract the signal from the noise. Nate Silver's here. Nate, we've been saying that since 2010, rip you off. Hey Marcus feeder. Oh, you have that trademarks. Okay. So anyway, welcome to the cube. You man who needs no introduction, but in case you don't know Nate, uh, he's a very famous author, five 30 eight.com. Statistician influence, influential individual predictor of a lot of things including presidential elections. And uh, great to have you here. Great to be here. So we listened to your keynote this morning. We asked earlier if some of our audience, can you tweet it and you know, what would you ask Nate silver? So of course we got the predictable, how the red Sox going to do this year? Who's going to be in the world series? Are we going to attack Syria? >>Uh, will the fed E's or tightened? Of course we're down here. Who'd you vote for? Or they, you know, they all want to know. And of course, a lot of these questions you can't answer because it's too far out. But, uh, but anyway, again, welcome, welcome to the cube. Um, so I want to start by, uh, picking up on some of the themes in your keynote. Uh, you're here at the Tableau conference. Obviously it's all about about data. Uh, and you, your basic, one of your basic premises was that, um, people will misinterpret data, they'll just use data for their own own biases. You have been a controversial figure, right? A lot of people have accused you of, of bias. Um, how, what do you F how do you feel about that as a person who's, uh, you know, statistician, somebody who loves data? >>I think everyone has bias in the sense that we all have one relatively narrow perspective as compared to a big set of problems that we all are trying to analyze or solve or understand together. Um, you know, but I do think some of this actually comes down to, uh, not just bias, but kind of personal morality and ethics really. It seems weird to talk about it that way, but there are a lot of people involved in the political world who are operating to manipulate public opinion, um, and that don't really place a lot of value on the truth. Right. And I consider that kind of immoral. Um, but people like that I think don't really understand that someone else might act morally by actually just trying to discover the way the objective world is and trying to use science and research to, to uncover things. >>And so I think it's hard people to, because if they were in your shoes, they would try and manipulate the forecast and they would cheat and put their finger on their scale. They assume that anyone else would do the same thing cause they, they don't own any. Yeah. So will you, you've made some incredibly accurate predictions, uh, in the face of, of, of others that clearly had bias that, that, that, you know mispredicted um, so how did you feel when you got those, those attacks? Were you flabbergasted? Were you pissed? Were you hurt? I mean, all of the above having you move houses for, for you? I mean you get used to them with a lot of bullshit, right? You're not too surprised. Um, I guess it surprised me how, but how much the people who you know are pretty intelligent are willing to, to fool themselves and how specious arguments where meet and by the way, people are always constructing arguments for, for outcomes they happen to be rooting for. >>Right? It'd be one thing if you said, well I'm a Republican, but boy I think Obama's going to crush Romney electoral college or vice versa. But you should have an extra layer of scrutiny when you have a view that diverges from the consensus or what kind of the markets are saying. And by the way, you can go and they're betting Margaret's, you can go and you could have bet on the outcome of election bookies in the UK, other countries. Right. And they kind of had forecast similar to ours. We were actually putting their money where their mouth was. Agree that Obama was a. Not a lot, but a pretty heavy favorite route. Most of the last two months in the election. I wanted to ask you about prediction markets cause as you probably know, I mean the betting public are actually very efficient. Handicappers right over. >>So I'll throw a two to one shot is going to be to three to one is going to be a four to one, you know, more often than not. But what are your thoughts on, on prediction markets? I mean you just sort of betting markets, you'd just alluded it to them just recently or is that a, is that a good, well there a lot there then then I think the punditry right. I mean, you know, so with, with prediction markets you have a couple of issues. Number one is do you have enough, uh, liquidity, um, and my volume in the markets for them to be, uh, uh, optimal. Right. And I think the answer right now is maybe not exactly. And like these in trade type markets, knowing trade has been, has been shut down. In fact, it was pretty light trading volumes. It might've had people who stood to gain or lose, um, you know, thousands of dollars. >>Whereas in quote, unquote real markets, uh, the stakes are, are several orders of magnitude higher. If you look at what happened to, for example, just prices of common stocks a day after the election last year, um, oil and gas stocks lost billions of dollars of market capitalization after Romney lost. Uh, conversely, some, you know, green tech stocks or certain types of healthcare socks at benefit from Obamacare going into play gain hundreds of millions, billions of dollars in market capitalization. So real investors have to price in these political risks. Um, anyway, I would love to have see fully legal, uh, trading markets in the U S people can get bet kind of proper sums of money where you have, um, a lot of real capital going in and people can kind of hedge their economic risk a little bit more. But you know, they're, they're bigger and it's very hard to beat markets. They're not flawless. And there's a whole chapter in the book about how, you know, the minute you assume that markets are, are clairvoyant and perfect, then that's when they start to fail. >>Ironically enough. But they're very good. They're very tough to beat and they certainly provide a reality check in terms of providing people with, with real incentives to actually, you know, make a bet on, on their beliefs and people when they have financial incentives, uh, uh, to be accurate then a lot of bullshit. There's a tax on bullshit is one way. That's okay. I've got to ask him for anyway that you're still a baseball fan, right? Is that an in Detroit fan? Right. I'm a tiger. There's my bias. You remember the bird? It's too young to remember a little too. I, so I grew up, I was born in 78, so 84, the Kirk Gibson, Alan Trammell teams are kind of my, my earliest. So you definitely don't remember Mickey Lola cha. I used to be a big guy. That's right fan as well. But so, but Sony, right when Moneyball came out, we just were at the Vertica conference. >>We saw Billy being there and, and uh, when, when, when, when, when that book came out, I said Billy Bean's out of his mind for releasing all these secrets. And you alluded to in your talk today that other teams like the rays and like the red Sox have sort of started to adopt those techniques. At the same time, I feel like culturally when another one of your V and your Venn diagram, I don't want you vectors, uh, that, that Oakland's done a better job of that, that others may S they still culturally so pushing back, even the red Sox themselves, it can be argued, you know, went out and sort of violated the, the principles were of course Oakland A's can't cause they don't have a, have a, have a budget to do. So what's your take on Moneyball? Is the, is the strategy that he put forth sustainable or is it all going to be sort of level playing field eventually? >>I mean, you know, the strategy in terms of Oh fine guys that take a lot of walks, right? Um, I mean everyone realizes that now it's a fairly basic conclusion and it was kind of the sign of, of how far behind how many biases there were in the market for that, you know, use LBP instead of day. And I actually like, but that, that was arbitrage, you know, five or 10 years ago now, um, put butts in the seat, right? Man, if they win, I guess it does, but even the red Sox are winning and nobody goes to the games anymore. The red Sox, tons of empty seats, even for Yankees games. Well, it's, I mean they're also charging 200 bucks a ticket or something. you can get a ticket for 20, 30 bucks. But, but you know, but I, you know, I, I, I mean, first of all, the most emotional connection to baseball is that if your team is in pennant races, wins world series, right then that produces multimillion dollar increases in ticket sales and, and TV contracts down the road. >>So, um, in fact, you know, I think one thing is, is looking at the financial side, like modeling the martial impact of a win, but also kind of modeling. If you do kind of sign a free agent, then, uh, that signaling effect, how much does that matter for season ticket sales? So you could do some more kind of high finance stuff in baseball. But, but some of the low hanging fruit, I mean, you know, almost every team now has a Cisco analyst on their payroll or increasingly the distinctions aren't even as relevant anymore. Right? Where someone who's first in analytics is also listening to what the Scouts say. And you have organizations that you know, aren't making these kind of distinctions between stat heads and Scouts at all. They all kind of get along and it's all, you know, finding better ways, more responsible ways to, to analyze data. >>And basically you have the advantage of a very clear way of measure, measure success where, you know, do you win? That's the bottom line. Or do you make money or, or both. You can isolate guys Marshall contribution. I mean, you know, I am in the process now of hiring a bunch of uh, writers and editors and developers for five 38 right? So someone has a column and they do really well. How much of that is on the, the writer versus the ed or versus the brand of the site versus the guy at ESPN who promoted it or whatever else. Right. That's hard to say. But in baseball, everyone kind of takes their turn. It's very easy to measure each player's kind of marginal contribution to sort of balance and equilibrium and, and, and it's potentially achieved. But, and again, from your talk this morning modeling or volume of data doesn't Trump modeling, right? >>You need both. And you need culture. You need, you need, you know, you need volume of data, you need high quality data. You need, uh, a culture that actually has the right incentives align where you really do want to find a way to build a better product to make more money. Right? And again, they'll seem like, Oh, you know, how difficult should it be for a company to want to make more money and build better products. But, um, when you have large organizations, you have a lot of people who are, uh, who are thinking very short term or only about only about their P and L and not how the whole company as a whole is doing or have, you know, hangups or personality conflicts or, or whatever else. So, you know, a lot of success I think in business. Um, and certainly when it comes to use of analytics, it's just stripping away the things that, that get in the way from understanding and distract you. >>It's not some wave a magic wand and have some formula where you uncover all the secrets in the world. It's more like if you can strip away the noise there and you're going to have a much clearer understanding of, of what's really there. Uh, Nate, again, thanks so much for joining us. So kind of wanna expand on that a little bit. So when people think of Nate silver, sometimes they, you know, they think Nate silver analytics big data, but you're actually a S some of your positions are kind of, you take issue with some of the core notions of big data really around the, the, the importance of causality versus correlation. So, um, so we had Kenneth kookier on from, uh, the economist who wrote a book about big data a while back, the strata conference. And you know, he, in that book, they talk a lot about it really doesn't matter how valid anymore, if you know that your customers are gonna buy more products based on this dataset or this correlation that it doesn't really matter why. >>You just try to try to try to exploit that. Uh, but in your book you talk about, well and in the keynote today you talked about, well actually hypothesis testing coming in with some questions and actually looking for that causality is also important. Um, so, so what is your, what is your opinion of kind of, you know, all this hype around big data? Um, you know, you mentioned volume is important, but it's not the only thing. I mean, like, I mean, I'll tell you I'm, I'm kind of an empiricist about anything, right? So, you know, if it's true that merely finding a lot of correlations and kind of very high volume data sets will improve productivity. And how come we've had, you know, kind of such slow economic growth over the past 10 years, where is the tangible increase in patent growth or, or different measures of progress. >>And obviously there's a lot of noise in that data set as well. But you know, partly why both in the presentation today and in the book I kind of opened up with the, with the history is saying, you know, let's really look at the history of technology. It's a kind of fascinating, an understudied feel, the link between technology and progress and growth. But, um, it doesn't always go as planned. And I certainly don't think we've seen any kind of paradigm shift as far as, you know, technological, economic productivity in the world today. I mean, the thing to remember too is that, uh, uh, technology is always growing in and developing and that if you have roughly 3% economic growth per year exponential, that's a lot of growth, right? It's not even a straight line growth. It's like exponential growth. And to have 3% exponential growth compounding over how many years is a lot. >>So you're always going to have new technologies developing. Um, but what I, I'm suspicious that as people will say this one technology is, is a game changer relative to the whole history of civilization up until now. Um, and also, you know, again, a lot of technologies you look at kind of economic models where you have different factors or productivity. It's not usually an additive relationship. It's more a multiplicative relationships. So if you have a lot of data, but people who aren't very good at analyzing it, you have a lot of data but it's unstructured and unscrutinised you know, you're not going to get particularly good results by and large. Um, so I just want to talk a little bit about the, the kind of the, the cultural issue of adopting kind of analytics and, and becoming a data driven organization. And you talk a lot about, um, you know, really what you do is, is setting, um, you know, try to predict the probabilities of something happening, not really predicting what's going to happen necessarily. >>And you talked to New York, you know, today about, you know, knowledging where, you know, you're not, you're not 100% sure acknowledging that this is, you know, this is our best estimate based on the data. Um, but of course in business, you know, a lot of people, a lot of, um, importance is put on kind of, you know, putting on that front that you're, you know, what you're talking about. It's, you know, you be confident, you go in, this is gonna happen. And, and sometimes that can actually move markets and move decision-making. Um, how do you balance that in a, in a business environment where, you know, you want to keep, be realistic, but you want to, you know, put forth a confident, uh, persona. Well, you know, I mean, first of all, everyone, I think the answer is that you have to, uh, uh, kind of take a long time to build the narrative correctly and kind of get back to the first principles. >>And so at five 38, it's kind of a case where you have a dialogue with the readers of the site every day, right? But it's not that you can solve in one conversation. If you come in to a boss who you never talked to you before, you have to present some PowerPoint and you're like, actually this initiative has a, you know, 57% chance of succeeding and the baseline is 50% and it's really good cause the upside's high, right? Like you know, that's going to be tricky if you don't have a good and open dialogue. And it's another barrier by the way to success is that uh, you know, none of this big data stuff is going to be a solution for companies that have poor corporate cultures where you have trouble communicating ideas where you don't everyone on the same page. Um, you know, you need buy in from, from all throughout the organization, which means both you need senior level people who, uh, who understand the value of analytics. >>You also need analysts or junior level people who understand what business problems the company is trying to solve, what organizational goals are. Um, so I mean, how do you communicate? It's tricky, you know, maybe if you can't communicate it, then you find another firm or go, uh, go trade stocks and, and uh, and short that company if you're not violating like insider trading rules of, of various kinds. Um, you know, I mean, the one thing that seems to work better is if you can, uh, depict things visually. People intuitively grasp uncertainty. If you kind of portray it to them in a graphic environment, especially with interactive graphics, uh, more than they might've just kind of put numbers on a page. You know, one thing we're thinking about doing with the new 580 ESPN, we're hiring a lot of designers and developers is in case where there is uncertainty, then you can press a button, kind of like a slot, Michigan and simulate and outcome many times, then it'll make sense to people. Right? And they do that already for, you know, NCAA tournament stuff or NFL playoffs. Um, but that can help. >>So Nate, I asked you my, my partner John furry, who's often or normally the cohost of this show, uh, just just tweeted me asking about crowd spotting. So he's got this notion that there's all this exhaust out there, the social exhaustive social data. How do you, or do you, or do you see the potential to use that exhaust that's thrown off from the connected consumer to actually make predictions? Um, so I'm >>a, I guess probably mildly pessimistic about this for the reason being that, uh, a lot of this data is very new and so we don't really have a way to kind of calibrate a model based on it. So you can look and say, well, you know, let's say Twitter during the Republican primaries in 2016 that, Oh, Paul Ryan is getting five times as much favorable Twitter sentiment as Rick Santorum or whatever among Republicans. But, but what's that mean? You know, to put something into a model, you have to have enough history generally, um, where you can translate X into Y by means of some function or some formula. And a lot of data is so new where you don't have enough history to do that. And the other thing too is that, um, um, the demographics of who is using social media is changing a lot. Where we are right now you come to conference like this and everyone has you know, all their different accounts but, but we're not quite there yet in terms of the broader population. >>Um, you have a lot of kind of thought leaders now a lot of, you know, kind of young, smart urban tech geeks and they're not necessarily as representative of the population as a whole. That will over time the data will become more valuable. But if you're kind of calibrating expectations based on the way that at Twitter or Facebook were used in 2013 to expect that to be reliable when you want a high degree of precision three years from now, even six months from now is, is I think a little optimistic. Some sentiment though, we would agree with that. I mean sentiment is this concept of how many people are talking about a thumbs up, thumbs down. But to the extent that you can get metadata and make it more stable, longer term, you would see potential there is, I mean, there are environments where the terrain is shifting so fast that by the time you know, the forecast that you'd be interested in, right? >>Like things have already changed enough where like it's hard to do, to make good forecast. Right? And I think one of the kind of fundamental themes here, one of my critiques is some of the, uh, of, uh, the more optimistic interpretations of big data is that fundamentally people are, are, most people want a shortcut, right? Most people are, are fairly lazy like labor. What's the hot stock? Yeah. Right. Um, and so I'm worried whenever people talk about, you know, biased interpretations of, of the data or information, right? Whenever people say, Oh, this is going to solve my problems, I don't have to work very hard. You know, not usually true. Even if you look at sports, even steroids, performance enhancing drugs, the guys who really get the benefits of the steroids, they have to work their butts off, right? And then you have a synergy which hell. >>So they are very free free meal tickets in life when they are going to be gobbled up in competitive environments. So you know, uh, bigger datasets, faster data sets are going to be very powerful for people who have the right expertise and the right partners. But, but it's not going to make, uh, you know anyone to be able to kind of quit their job and go on the beach and sip my ties. So ne what are you working on these days as it relates to data? What's exciting you? Um, so with the, with the move to ESPN, I'm thinking more about, uh, you know, working with them on sports type projects, which is something having mostly cover politics. The past four or five years I've, I've kind of a lot of pent up ideas. So you know, looking at things in basketball for example, you have a team of five players and solving the problem of, of who takes the shot, when is the guy taking a good shot? >>Cause the shot clock's running out. When does a guy stealing a better opportunity from, from one of his teammates. Question. We want to look at, um, you know, we have the world cup the summer, so soccer is an interest of mine and we worked in 2010 with ESPN on something called the soccer power index. So continuing to improve that and roll that out. Um, you know, obviously baseball is very analytics rich as well, but you know, my near term focus might be on some of these sports projects. Yeah. So that the, I have to ask you a followup on the, on the soccer question. Is that an individual level? Is that a team level of both? So what we do is kind of uh, uh, one problem you have with the national teams, the Italian national team or Brazilian or the U S team is that they shift their personnel a lot. >>So they'll use certain guys for unimportant friendly matches for training matches that weren't actually playing in Brazil next year. So the system soccer power next we developed for ESPN actually it looks at the rosters and tries to make inferences about who is the a team so to speak and how much quality improvement do you have with them versus versus, uh, guys that are playing only in the marginal and important games. Okay. So you're able to mix and match teams and sort of predict on your flow state also from club league play to make inferences about how the national teams will come together. Um, but soccer is a case where, where we're going into here where we had a lot more data than we used to. Basically you had goals and bookings, I mean, and yellow cards and red cards and now you've collected a lot more data on how guys are moving throughout the field and how many passes there are, how much territory they're covering, uh, tackles and everything else. So that's becoming a lot smarter. Excellent. All right, Nate, I know you've got to go. I really appreciate the time. Thanks for coming on. The cube was a pleasure to meet you. Great. Thank you guys. All right. Keep it right there, everybody. We'll be back with our next guest. Dave Volante and Jeff Kelly. We're live at the Tableau user conference. This is the cube.