>> Hello, everyone. Welcome to theCUBE's coverage of "Compute Engineered for your Hybrid World," sponsored by HPE and Intel. Today we're going to discuss the new 4th Gen Intel Xeon Scalable process impact on containers and AI. I'm John Furrier, your host of theCUBE, and I'm joined by three experts to guide us along. We have Jordan Plum, Senior Director of AI and products for Intel, Bradley Sweeney, Big Data and AI Product Manager, Mainstream Compute Workloads at HPE, and Gary Wang, Containers Product Manager, Mainstream Compute Workloads at HPE. Welcome to the program gentlemen. Thanks for coming on. >> Thanks John. >> Thank you for having us. >> This segment is going to be talking about containers to deploy high performance AI applications. This is a really important area right now. We're seeing a lot more AI deployed, kind of next gen AI coming. How is HPE supporting and testing and delivering containers for AI? >> Yeah, so what we're doing from HPE's perspective is we're taking these container platforms, combining with the next generation Intel servers to fully validate the deployment of the containers. So what we're doing is we're publishing the reference architectures. We're creating these automation scripts, and also creating a monitoring and security strategy for these container platforms. So for customers to easily deploy these Kubernete clusters and to easily secure their community environments. >> Gary, give us a quick overview of the new Proliant DL 360 and 380 Gen 11 servers. >> Yeah, the load, for example, for container platforms what we're seeing mostly is the DL 360 and DL 380 for matching really well for container use cases, especially for AI. The DL 360, with the expended now the DDR five memory and the new PCI five slots really, really helps the speeds to deploy these container environments and also to grow the data that's required to store it within these container environments. So for example, like the DL 380 if you want to deploy a data fabric whether it's the Ezmeral data fabric or different vendors data fabric software you can do so with the DL 360 and DL 380 with the new Intel Xeon processors. >> How does HP help customers with Kubernetes deployments? >> Yeah, like I mentioned earlier so we do a full validation to ensure the container deployment is easy and it's fast. So we create these automation scripts and then we publish them on GitHub for customers to use and to reference. So they can take that and then they can adjust as they need to. But following the deployment guide that we provide will make the, deploy the community deployment much easier, much faster. So we also have demo videos that's also published and then for reference architecture document that's published to guide the customer step by step through the process. >> Great stuff. Thanks everyone. We'll be going to take a quick break here and come back. We're going to do a deep dive on the fourth gen Intel Xeon scalable process and the impact on AI and containers. You're watching theCUBE, the leader in tech coverage. We'll be right back. (intense music) Hey, welcome back to theCUBE's continuing coverage of "Compute Engineered for your Hybrid World" series. I'm John Furrier with the Cube, joined by Jordan Plum with Intel, Bradley Sweeney with HPE, and Gary Wang from HPE. We're going to do a drill down and do a deeper dive into the AI containers with the fourth gen Intel Xeon scalable processors we appreciate your time coming in. Jordan, great to see you. I got to ask you right out of the gate, what is the view right now in terms of Intel's approach to containers for AI? It's hot right now. AI is booming. You're seeing kind of next gen use cases. What's your approach to containers relative to AI? >> Thanks John and thanks for the question. With the fourth generation Xeon scalable processor launch we have tested and validated this platform with over 400 deep learning and machine learning models and workloads. These models and workloads are publicly available in the framework repositories and they can be downloaded by anybody. Yet customers are not only looking for model validation they're looking for model performance and performance is usually a combination of a given throughput at a target latency. And to do that in the data center all the way to the factory floor, this is not always delivered from these generic proxy models that are publicly available in the industry. >> You know, performance is critical. We're seeing more and more developers saying, "Hey, I want to go faster on a better platform, faster all the time." No one wants to run slower stuff, that's for sure. Can you talk more about the different container approaches Intel is pursuing? >> Sure. First our approach is to meet the customers where they are and help them build and deploy AI everywhere. Some customers just want to focus on deployment they have more mature use cases, and they just want to download a model that works that's high performing and run. Others are really focused more on development and innovation. They want to build and train models from scratch or at least highly customize them. Therefore we have several container approaches to accelerate the customer's time to solution and help them meet their business SLA along their AI journey. >> So what developers can just download these containers and just go? >> Yeah, so let me talk about the different kinds of containers we have. We start off with pre-trained containers. We'll have about 55 or more of these containers where the model is actually pre-trained, highly performant, some are optimized for low latency, others are optimized for throughput and the customers can just download these from Intel's website or from HPE and they can just go into production right away. >> That's great. A lot of choice. People can just get jump right in. That's awesome. Good, good choice for developers. They want more faster velocity. We know that. What else does Intel provide? Can you share some thoughts there? What you guys else provide developers? >> Yeah, so we talked about how hey some are just focused on deployment and they maybe they have more mature use cases. Other customers really want to do some more customization or optimization. So we have another class of containers called development containers and this includes not just the kind of a model itself but it's integrated with the framework and some other capabilities and techniques like model serving. So now that customers can download just not only the model but an entire AI stack and they can be sort of do some optimizations but they can also be sure that Intel has optimized that specific stack on top of the HPE servers. >> So it sounds simple to just get started using the DL model and containers. Is that it? Where, what else are customers looking for? What can you take a little bit deeper? >> Yeah, not quite. Well, while the customer customer's ability to reproduce performance on their site that HPE and Intel have measured in our own labs is fantastic. That's not actually what the customer is only trying to do. They're actually building very complex end-to-end AI pipelines, okay? And a lot of data scientists are really good at building models, really good at building algorithms but they're less experienced in building end-to-end pipelines especially 'cause the number of use cases end-to-end are kind of infinite. So we are building end-to-end pipeline containers for use cases like media analytics and sentiment analysis, anomaly detection. Therefore a customer can download these end-to-end containers, right? They can either use them as a reference, just like, see how we built them and maybe they have some changes in their own data center where they like to use different tools, but they can just see, "Okay this is what's possible with an end-to-end container on top of an HPE server." And other cases they could actually, if the overlap in the use case is pretty close, they can just take our containers and go directly into production. So this provides developers, all three types of containers that I discussed provide developers an easy starting point to get them up and running quickly and make them productive. And that's a really important point. You talked a lot about performance, John. But really when we talk to data scientists what they really want to be is productive, right? They're under pressure to change the business to transform the business and containers is a great way to get started fast >> People take product productivity, you know, seriously now with developer productivity is the hottest trend obviously they want performance. Totally nailed it. Where can customers get these containers? >> Right. Great, thank you John. Our pre-trained model containers, our developmental containers, and our end-to-end containers are available at intel.com at the developer catalog. But we'd also post these on many third party marketplaces that other people like to pull containers from. And they're frequently updated. >> Love the developer productivity angle. Great stuff. We've still got more to discuss with Jordan, Bradley, and Gary. We're going to take a short break here. You're watching theCUBE, the leader in high tech coverage. We'll be right back. (intense music) Welcome back to theCUBE's coverage of "Compute Engineered for your Hybrid World." I'm John Furrier with theCUBE and we'll be discussing and wrapping up our discussion on containers to deploy high performance AI. This is a great segment on really a lot of demand for AI and the applications involved. And we got the fourth gen Intel Xeon scalable processors with HP Gen 11 servers. Bradley, what is the top AI use case that Gen 11 HP Proliant servers are optimized for? >> Yeah, thanks John. I would have to say intelligent video analytics. It's a use case that's supplied across industries and verticals. For example, a smart hospital solution that we conducted with Nvidia and Artisight in our previous customer success we've seen 5% more hospital procedures, a 16 times return on investment using operating room coordination. With that IVA, so with the Gen 11 DL 380 that we provide using the the Intel four gen Xeon processors it can really support workloads at scale. Whether that is a smart hospital solution whether that's manufacturing at the edge security camera integration, we can do it all with Intel. >> You know what's really great about AI right now you're starting to see people starting to figure out kind of where the value is does a lot of the heavy lifting on setting things up to make humans more productive. This has been clearly now kind of going neck level. You're seeing it all in the media now and all these new tools coming out. How does HPE make it easier for customers to manage their AI workloads? I imagine there's going to be a surge in demand. How are you guys making it easier to manage their AI workloads? >> Well, I would say the biggest way we do this is through GreenLake, which is our IT as a service model. So customers deploying AI workloads can get fully-managed services to optimize not only their operations but also their spending and the cost that they're putting towards it. In addition to that we have our Gen 11 reliance servers equipped with iLO 6 technology. What this does is allows customers to securely manage their server complete environment from anywhere in the world remotely. >> Any last thoughts or message on the overall fourth gen intel Xeon based Proliant Gen 11 servers? How they will improve workload performance? >> You know, with this generation, obviously the performance is only getting ramped up as the needs and requirements for customers grow. We partner with Intel to support that. >> Jordan, gimme the last word on the container's effect on AI applications. Your thoughts as we close out. >> Yeah, great. I think it's important to remember that containers themselves don't deliver performance, right? The AI stack is a very complex set of software that's compiled together and what we're doing together is to make it easier for customers to get access to that software, to make sure it all works well together and that it can be easily installed and run on sort of a cloud native infrastructure that's hosted by HPE Proliant servers. Hence the title of this talk. How to use Containers to Deploy High Performance AI Applications. Thank you. >> Gentlemen. Thank you for your time on the Compute Engineered for your Hybrid World sponsored by HPE and Intel. Again, I love this segment for AI applications Containers to Deploy Higher Performance. This is a great topic. Thanks for your time. >> Thank you. >> Thanks John. >> Okay, I'm John. We'll be back with more coverage. See you soon. (soft music)

>>from around the globe. It's the Cube with digital coverage of Dell Technologies. World Digital Experience Brought to You by Dell Technologies. >>Welcome to the Cubes Coverage of Dell Technologies World 2020. The Digital Experience. I'm Lisa Martin, and I'm pleased to welcome back a Cube alumni and a new Cube member to the program today. China. My Mondal is back with US Director of Solutions Marketing for Dell Technologies China. But it's great to see you at Dell Technologies world, even though we're very specially death. >>Happy to be back. Thank you, Lisa. >>And Thomas Henson is joining us for the first time. Global business development manager for a I and analytics. Thomas, Welcome to the Cube. >>I am excited to be here. It's my first virtual cube. >>Yeah, well, you better make it a good one. All right. I said we're talking about a I so so much has changed John to me. The last time I saw you were probably were sitting a lot closer together. So much has changed in the last 67 months, but a lot has changed with the adoption of Ai Thomas. Kick us off. What are some of the big things feeling ai adoption right now? >>Yeah, I >>would have to >>say the two biggest things right now or as we look at accelerated compute and by accelerated compute we're not just talking about the continuation of Moore's law, but how In Data Analytics, we're actually doing more processing now with GP use, which give us faster insights. And so now we have the ability to get quicker insights in jobs that may have taken, you know, taking weeks to months a song as we were measuring. And then the second portion is when we start to talk about the innovation going on in the software and framework world, right? So no longer do you have toe know C plus plus or a lower level language. You can actually do it in Python and even pull it off of Get Hub. And it's all part of that open source community. So we're seeing Mawr more folks in the field of data science and deep learning that can actually implement some code. And then we've got faster compute to be able to process that. >>Tell me, what are your thoughts? >>Think I want to add? Is the explosive growth off data on that's actually are fulfilling the AI adoption. Think off. Like all the devices we have, the i o t. On age devices are doing data are pumping data into the pipeline. Our high resolution satellite imagery, all social media generating data. No. All of this data are actually helping the adoption off a I because now we have very granular data tow our friend the AI model Make the AI models are much better. Besides, so the combination off both in, uh, data the power off Like GPU, power surfers are coupled with the inefficient in the eye after and tools helping off. Well, the AI growth that we're seeing today >>trying to make one of the things that we've known for a while now is that it's for a I to be valuable. It's about extracting value from that. Did it? You talked about the massive explosion and data, but yet we know for a long time we've been talking about AI for decades. Initiatives can fail. What can Dell Technologies do now to help companies have successfully I project? >>Yeah, eso As you were saying, Lisa, what we're seeing is the companies are trying to add up AI Technologies toe Dr Value and extract value from their data set. Now the way it needs to be framed is there is a business challenge that customers air trying to solve. The business challenge gets transformed into a data science problem. That data scientist is going toe work with the high technology, trained them on it. That data science problem gets to the data science solution on. Then it needs to be mapped to production deployment as a business solution. What happens? Ah, lot off. The time is the companies do not plan for output transition from all scale proof of concept that it a scientists are playing with, like a smaller set of data two, when it goes toe the large production deployment dealing with terabytes toe terabyte self data. Now that's where we come in. At their technologies, we have into end solutions for the, uh for the ai for pollution in the customers journeys starting from proof of concept to production. And it is all a seamless consular and very scalable. >>So if some of the challenges there are just starting with iterations. Thomas question for you as business development manager, those folks that John um I talked about the data scientists, the business. How are you helping them come together from the beginning so that when the POC is initiated, it actually can go on the right trajectory to be successful? >>No, that's a great point. And just to kind of build off of what Shonda my was talking about, You know, we call it that last mile, right? Like, Hey, I've got a great POC. How do I get into production? Well, if you have executive sponsorship and it's like, Hey, everybody was on board, but it's gonna take six months to a year. It's like, Whoa, you're gonna lose some momentum. So where we help our customers is, you know, by partnering with them to show them how to build, you know, from an i t. And infrastructure perspective what that ai architectural looks like, right? So we have multiple solutions around that, and at the end of the day, it's about just like Sean. Um, I was saying, You know, we may start off with a project that maybe it's only half a terabyte. Maybe it's 10 terabytes, but once you go into production, if it turns out to be three petabytes four petabytes. Nobody really, you know, has the infrastructure built unless they built on those solid practices. And that's where our solutions come in. So we can go from small scale laboratory all the way large scale production without having to move any of that data. Right? So, you know, at the heart of that is power scale and giving you that ability to scale your data and no more data migration so that you can handle one PC or multiple PCs as those models continue to improve as you start to move into production >>and I'm sticking with you 1st. 2nd 0, sorry. Trying to go ahead. >>So I was going to add that, uh, just like posthumous said right. So if you were a data scientist, you are working with this data science workstations, but getting the data from, uh, L M c our scales thes scale out platform and, uh, as it is growing from, you see two large kills production data can stay in place with the power scale platform. You can add notes, and it can grow to petabytes. And you can add in not just the workstations, but also our They'll power it, solve our switches building out our enter A I ready solutions are already solution for your production. Giving are very seamless experience from the data scientist with the i t. >>So China may will stick with you then. I'm curious to know in the last 6 to 7 months since 2020 has gone in a very different direction thing we all would have predicted our last Dell Technologies world together. What are you seeing? China. My in terms of acceleration or maybe different industries. What our customers needs, how they changed. I guess I should say in the in 2020. >>So in 2020 we're seeing the adoption off a I even more rapidly. Uh, if you think about customers ranging from like say, uh, media and entertainment industry toe, uh, the customer services off any organization to, uh the healthcare and life sciences with lots off genome analysts is going on in all of these places where we're dealing with large are datasets. We're seeing ah, lot off adoption foster processing off A. I R. Technologies, uh, giving with, say, the all the research that the's Biosciences organizations are happening. Uh, Thomas, I know like you are working with, like, a customer. So, uh, can you give us a little bit more example in there? >>Yes, one of the areas. You know, we're talking about 2021 of the things that we're seeing Mawr and Mawr is just the expansion of Just look at the need for customer support, right arm or folks working remotely their arm or folks that are learning remote. I know my child is going through virtual schools, So think about your I t organization and how Maney calls you're having now to expand. And so this is a great area where we're starting to see innovation within a I and model building to be ableto have you know, let's call it, you know, the next generation of chatbots rights. You can actually build these models off the data toe, augment those soup sports systems >>because you >>have two choices, right? You can either. You know, you you can either expand out your call center right for for we're not sure how long or you can use AI and analytics to help augment to help maybe answer some of those first baseline questions. The great thing about customers who are choosing power scale and Dell Technologies. Their partner is they already have. The resource is to be able to hold on to that data That's gonna help them train those models to help. >>So, Thomas, whenever we're talking about data, the explosions it brings to mind compliance. Protection, security. We've seen ransom where really skyrocket in 2020. Just you know, the other week there was the VA was hit. Um, I think there was also a social media Facebook instagram ticktock, 235 million users because there was an unsecured cloud database. So that vector is expanding. How can you help customers? Customers accelerate their AI projects? Well, ensuring compliance and protection and security of that data. >>Really? That's the sweet spot for power scale. We're talking with customers, right? You know, built on one FS with all the security features in mind. And I, too, came from the analytics world. So I remember in the early days of Hadoop, where, you know, as a software developer, we didn't need security, right? We you know, we were doing researching stuff, but then when we took it to the customer and and we're pushing to production, But what about all the security features. We needed >>the same thing >>for artificial intelligence, right? We want toe. We want to make sure that we're putting those security features and compliance is in. And that's where you know, from from an AI architecture perspective, by starting with one FS is at the heart of that solution. You can know that you're protecting for you know, all the enterprise features that you need, whether it be from compliance, thio, data strategy, toe backup and recovery as well. >>So when we're talking about big data volumes Chanda, mind we have to talk about the hyper scale er's talk to us about, you know, they each offer azure A W s Google cloud hundreds of AI services. So how does DEL help customers use the public cloud the data that's created outside of it and use all of those use that the right AI services to extract that value? >>Yeah. Now, as you mentioned, all of these hyper scholars are they differentiate with our office is like a i m l r Deep Learning Technologies, right? And as our customer, you want toe leverage based off all the, uh, all the cloud has to offer and not stuck with one particular cloud provider. However, we're talking about terabytes off data, right? So if you are happy with what doing service A from cloud provider say Google what you want to move to take advantage off another surface off from Asia? It comes with a very high English p a migration risk on time it will take to move the data itself. Now that's not good, right? As the customer, we should be able to live for it. Best off breed our cloud services for AI and for that matter, for anything across the board. Now, how we help customers is you can have all of your data say, in a managed, uh, managed cloud service provider running on power scale. But then you can connect from this managed cloud service provider directly toe any off the hyper scholars. You can connect toe aws, azure, Google Cloud and even, like even, uh, the in place analytics that power scale offers you can run. Uh, those, uh I mean, run those clouds AI services directly on that data simultaneously from these three, and I'll add like one more thing, right? Thes keep learning. Technologies need GPU power solvers, right? and cloud even within like one cloud is not homogeneous environment. Like sometimes you'll find a US East has or gp part solvers. But like you are in the West and the same for other providers. No, with our still our technologies cloud power scale for multi cloud our scale is sitting outside off those hyper scholars connected directly to our any off this on. Then you can burst into different clouds, take advantage off our spot. Instances on are like leverage. All the GP is not from one particular service provider part. All of those be our hyper scholars. So those are some examples off the work we're doing in the multi cloud world for a I >>So that's day. You're talking about data there. So powers failed for multi cloud for data that's created outside the public club. But Thomas, what about for data that's created inside the cloud? How does Del help with that? >>Yes. So, this year, we actually released a solution, uh, in conjunction with G C. P. So within Google Cloud, you can have power scale for one fs, right? And so that's that native native feature. So, you know, goes through all the compliance and all the features within being a part of that G c p natively eso counts towards your credits and your GP Google building as well. But it's still all the features that you have. And so we've been running some, actually, some benchmarks. So we've got a couple of white papers out there, that kind of detail. You know what we can do from an artificial intelligence perspective back to Sean Demise Example. We were just talking about, you know, being able to use more and more GPU. So we we've done that to run some of our AI benchmarks against that and then also, you know, jumped into the Hadoop space. But because you know, that's 11 area from a power scale, prospective customers were really interested. Um, and they have been for years. And then, really, the the awesome portion about this is for customers that are looking for a hybrid solution. Or maybe it's their first kickoff to it. So back Lisa to those compliance features that we were talking about those air still inherent within that native Google G C P one fs version, but then also for customers that have it on prim. You can use those same features to burst your data into, um, your isil on cluster using all the same native tools that you've been using for years within your enterprise. >>God, it's so starting out for power. Skill for Google Cloud Trying to get back to you Kind of wrapping things up here. What are some of the things that we're going to see next from Dell from an AI Solutions perspective? >>Yes. So we are working on many different interesting projects ranging from, uh, the latest, uh, in video Salford's that they have announced d d x a 100. And in fact, two weeks ago at GTC, uh, Syria announced take too far parts with, uh, it takes a 100 solvers. We're part off that ecosystem. And we are working with, uh, the leading, uh uh, solutions toe benchmark, our ai, uh, environments, uh, for all the storage, uh, ensuring, like we are providing, like, all the throughput and scalability that we have to offer >>Thomas finishing with you from the customer perspective. As we talked about so many changes this year alone as we approach calendar year 2021 what are some of the things that Dell is doing with its customers with its partners, the hyper scale er's and video, for example, Do you think customers are really going to be able to truly accelerate successful AI projects? >>Yeah. So the first thing I'd like to talk about is what we're doing with the D. G. S A 100. So this month that GTC you saw our solution for a reference architecture for the G s, a 100 plus power scale. So you talk about speed and how we can move customers insights. I mean, some of the numbers that we're seeing off of that are really a really amazing right. And so this is gives the customers the ability to still, you know, take all the features and use use I salon and one f s, um, like they have in the past, but now combined with the speed of the A 100 still be ableto speed up. How fast they're using those building out those deep learning models and then secondly, with that that gives them the ability to scale to. So there's some features inherent within this reference architecture that allow for you to make more use, right? So bring mawr data scientists and more modelers GP use because that's one thing you don't see Data scientist turning away right there always like, Hey, you know, I mean, this this project here needs needs a GPU. And so, you know, from a power scale one fs perspective, we want to be able to make sure that we're supporting that. So that as that data continues to grow, which, you know we're seeing is one of the large factors. Whenever we're talking about artificial intelligence is the scale for the data. We wanna them to be able to continue to build out that data consolidation area for all these multiple different workloads. That air coming in. >>Excellent, Thomas. Thanks for sharing that. Hopefully next time we get to see you guys in person and we can talk about a customer who has done something very successful with you guys. Kind of me. Always great to talk to you. Thank you for joining us. >>Thank you. Thank you >>for China. May Mandel and Thomas Henson. I'm Lisa Martin. You're watching the cubes Coverage of Dell Technologies, World 2020

>> Live from San Francisco, California, it's theCUBE, covering the IBM Chief Data Officer Summit. Brought to you by IBM. >> We're back at San Francisco. We're here at Fisherman's Wharf covering the IBM Chief Data Officer event #IBMCDO. This is the tenth year of this event. They tend to bookend them both in San Francisco and in Boston, and you're watching theCUBE, the leader in live tech coverage. My name is Dave Valante. John Thomas is here, Cube alum and distinguished engineer, Director of Analytics at IBM, and somebody who provides technical direction to the data science elite team. John, good to see you again. Steve Aliouk is back. He is the Vice President of Deep Learning in the Global Chief Data Office, thanks for comin' on again. >> No problem. >> Let's get into it. So John, you and I have talked over the years at this event. What's new these days, what are you working on? >> So Dave, still working with clients on implementing data science and AI data use cases, mostly enterprise clients, and seeing a variety of different things developing in that space. Things have moved into broader discussions around AI and how to actually get value out of that. >> Okay, so I know one of the things that you've talked about is operationalizing machine intelligence and AI and cognitive and that's always a challenge, right. Sounds good, we see this potential but unless you change the operating model, you're not going to get the type of business value, so how do you operationalize AI? >> Yeah, this is a good question Dave. So, enterprises, many of them, are beginning to realize that it is not enough to focus on just the coding and development of the models, right. So they can hire super-talented Python TensorFlow programmers and get the model building done, but there's no value in it until these models actually are operationalized in the context of the business. So one aspect of this is, actually we know, we are thinking of this in a very systematic way and talking about this in a prescriptive way. So, you've got to scope your use cases out. You got to understand what is involved in implementing the use case. Then the steps are build, run, manage, and each of these have technical aspects and business aspects around, right. So most people jump right into the build aspect, which is writing the code. Yeah, that's great, but once you build the code, build the models by writing code, how do you actually deploy these models? Whether that is for online invocation or back storing or whatever, how do you manage the performance of these models over time, how do you retrain these models, and most importantly, when these models are in production, how do I actually understand the business metrics around them? 'Cause this goes back to that first step of scoping. What are the business KPI's that the line of business cares about? The data scientist talks about data science metrics, position and recall and Area Under the ROC Curve and accuracy and so on. But how do these relate to business KPI's. >> All right, so we're going to get into each of those steps in a moment, but Steve I want to ask you, so part of your charter, Inderpal, Global Chief Data Officer, you guys have to do this for IBM, right, drink your own champagne, dog footing, whatever you call it. But there's real business reasons for you to do that. So how is IBM operationalizing AI? What kind of learnings can you share? >> Well, the beauty is I got a wide portfolio of products that I can pull from, so that's nice. Like things like AI open to Watson, some of the hardware components, all that stuffs kind of being baked in. But part of the reason that John and I want to do this interview together, is because what he's producing, what his thoughts are kind of resonates very well for our own practices internally. We've got so many enterprise use cases, how are we deciding, you know, which ones to work on, which ones have the data, potentially which ones have the biggest business impact, all those KPI's etcetera, also, in addition to, for the practitioners, once we decide on a specific enterprise use case to work on, when have they reached the level where the enterprise is having a return on investment? They don't need to keep refining and refining and refining, or maybe they do, but they don't know these practitioners. So we have to clearly justify it, and scope it accordingly, or these practitioners are left in this kind of limbo, where they're producing things, but not able to iterate effectively for the business, right? So that process is a big problem I'm facing internally. We got hundreds of internal use cases, and we're trying to iterate through them. There's an immense amount of scoping, understanding, etcetera, but at the same time, we're building more and more technical debt, as the process evolves, being able to move from project to project, my team is ballooning, we can't do this, we can't keep growing, they're not going to give me another hundred head count, another hundred head count, so we're definitely need to manage it more appropriately. And that's where this mentality comes in there's-- >> All right, so I got a lot of questions. I want to start unpacking this stuff. So the scope piece, that's we're setting goals, identifying the metrics, success metrics, KPI's, and the like, okay, reasonable starting point. But then you go into this, I think you call it, the explore or understanding phase. What's that all about, is that where governance comes in? >> That's exactly where governance comes in. Right, so because it is, you know, we all know the expression, garbage in, garbage out, if you don't know what data you're working with for your machine learning and deep learning enterprise projects, you will not have the resource that you want. And you might think this is obvious, but in an enterprise setting, understanding where the data comes from, who owns the data, who work on the data, the lineage of that data, who is allowed access to the data, policies and rules around that, it's all important. Because without all of these things in place, the models will be questioned later on, and the value of the models will not realized, right? So that part of exploration or understanding, whatever you want to call it, is about understanding data that has to be used by the ML process, but then at a point in time, the models themselves need to be cataloged, need to be published, because the business as a whole needs to understand what models have been produced out of this data. So who built these models? Just as you have lineage of data, you need lineage of models. You need to understand what API's are associated with the models that are being produced. What are the business KPI's that are linked to model metrics? So all of that is part of this understand and explore path. >> Okay, and then you go to build. I think people understand that, everybody wants to start there, just start the dessert, and then you get into the sort of run and manage piece. Run, you want a time to value, and then when you get to the management phase, you really want to be efficient, cost-effective, and then iterative. Okay, so here's the hard question here is. What you just described, some of the folks, particularly the builders are going to say, "Aw, such a waterfall approach. Just start coding." Remember 15 years ago, it was like, "Okay, how do we "write better software, just start building! "Forget about the requirements, "Just start writing code." Okay, but then what happens, is you have to bolt on governance and security and everything else so, talk about how you are able to maintain agility in this model. >> Yeah, I was going to use the word agile, right? So even in each of these phases, it is an agile approach. So the mindset is about agile sprints and our two week long sprints, with very specific metrics at the end of each sprint that is validated against the line of business requirements. So although it might sound waterfall, you're actually taking an agile approach to each of these steps. And if you are going through this, you have also the option to course correct as it goes along, because think of this, the first step was scoping. The line of business gave you a bunch of business metrics or business KPI's they care about, but somewhere in the build phase, past sprint one or sprint 2, you realize, oh well, you know what, that business KPI is not directly achievable or it needs to be refined or tweaked. And there is that circle back with the line of business and a course correction as it was. So it's a very agile approach that you have to take. >> Are they, are they, That's I think right on, because again, if you go and bolt on compliance and governance and security after the fact, we know from years of experience, that it really doesn't work well. You build up technical debt faster. But are these quasi-parallel? I mean there's somethings that you can do in build as the scoping is going on. Is there collaboration so you can describe, can you describe that a little bit? >> Absolutely, so for example, if I know the domain of the problem, I can actually get started with templates that help me accelerate the build process. So I think in your group, for example, IBM internally, there are many, many templates these guys are using. Want to talk a little bit about that? >> Well, we can't just start building up every single time. You know, that's again, I'm going to use this word and really resonate it, you know it's not extensible. Each project, we have to get to the point of using templates, so we had to look at those initiatives and invest in those initiatives, 'cause initially it's harder. But at least once we have some of those cookie-cutter templates and some of them, they might have to have abstractions around certain parts of them, but that's the only way we're ever able to kind of tackle so many problems. So no, without a doubt, it's an important consideration, but at the same time, you have to appreciate there's a lot of projects that are fundamentally different. And that's when you have to have very senior people kind of looking at how to abstract those templates to make them reusable and consumable by others. >> But the team structure, it's not a single amoeba going through all these steps right? These are smaller teams that are, and then there's some threading between each step? >> This is important. >> Yeah, that's tough. We were just talking about that concept. >> Just talking about skills and >> The bind between those groups is something that we're trying to figure out how to break down. 'Cause that's something he recognizes, I recognize internally, but understanding that those peoples tasks, they're never going to be able to iterate through different enterprise problems, unless they break down those borders and really invest in the communication and building those tools. >> Exactly, you talk about full stack teams. So you, it is not enough to have coding skills obviously. >> Right. What is the skill needed to get this into a run environment, right? What is the skill needed to take metrics like not metrics, but explainability, fairness in the moderates, and map that to business metrics. That's a very different skill from Python coding skills. So full stack teams are important, and at the beginning of this process where someone, line of business throws 100 different ideas at you, and you have to go through the scoping exercise, that is a very specific skill that is needed, working together with your coders and runtime administrators. Because how do you define the business KPI's and how do you refine them later on in the life cycle? And how do you translate between line of business lingo and what the coders are going to call it? So it's a full stack team concept. It may not necessarily all be in one group, it may be, but they have to work together across these different side loads to make it successful. >> All right guys, we got to leave it there, the trains are backing up here at IBM CDO conference. Thanks so much for sharing the perspectives on this. All right, keep it right there everybody. You're watchin' "theCUBE" from San Francisco, we're here at Fisherman's Wharf. The IBM Chief Data Officer event. Right back. (bubbly electronic music)

>> Live from San Francisco, California, it's the Cube, covering the IBM Chief Data Officer Summit, brought to you by IBM. >> Hello, everyone. Welcome to historic Fisherman's Wharf in San Francisco. We're covering the IBM Chief Data Officer event, #IBMCDO. This is the Cube's, I think, eighth time covering this event. This is the tenth year anniversary of the IBM CDO event, and it's a little different format today. We're here at day one. It's like a half day. They start at noon, and then the keynotes. We're starting a little bit early. We're going to go all day today. My name is Dave Volante. Steve Eliuk is here. He's a Cube alum and Vice President of Deep Learning and the Global Chief Data Officer at IBM. And Tim Humphrey, the VP at the Chief Data Office at IBM. Gents, welcome to the Cube. >> Welcome, glad to be here. >> So, couple years ago, Ginni Rometty, at a big conference, talked about incumbent disruptors, and the whole notion was that you've got established businesses that need to transform into data businesses. Well, that struck me, that well, if IBM's going to sell that to its customers, it has to go through its own transformation, Steve. So let's start there. What is IBM doing to transform into a data company? >> Well, I've been at IBM for, you know, two years now, and luckily I'm benefiting from a lot of that transformation that's taken place over the past three or four years. So, internally, getting (mumbling) in order, understanding it, going through various different foundation stones, building those building blocks so that we can gather new insights and traverse through the cognitive journey. One of the nice things though, is that we have such a wide, diverse set of data within the company. So for different types of enterprise use cases that have benefits from AI, we have a lot of data assets that we can pull from. Now, keeping those data assets in good order is a challenging task in itself. And I'm able to pull from a lot of different tools that IBM's building for our customers. I get to use them internally, look at them, evaluate them, give them real practitioner's point of view to ultimately get insight for our internal business practices, but also for our customers in turn. >> Okay, so, when you think about a data business, they've got data at the core. I'm going to draw a, like, simple conceptual picture, and you've got people around it, maybe you've got processes around it. IBM, hundred-plus-year-old company, you've got different things at the core. It's products. It's people. It's business process. So maybe you could talk, Tim, about how you guys have gone about putting data at the center of the universe. Is that the right way to think about it? >> It is the right way to think about it, and I like how you were describing it. Because when you think about IBM, we've been around over a hundred years, and we do business in roughly over 170 countries. And we have businesses that span hardware, software, services, financing. And along the way, we've also acquired and divested a lot of companies and a lot of businesses. So what that leaves you with is a very fragmented data landscape, right? You know, to support regulations in this country, taxes, tax rules in another country, and having all these different types of businesses. Some you inherit. Some are born from within your company. It just leaves a lot of data silos. And as we see transformations being so important, and data is at the heart of that transformation, it was important for us to really be able to organize ourselves such that access to data is not a problem. Such that being able to combine data across disciplines from finance to HR to sales to marketing to procurement. That was the big challenge, right? And to do this in a way that really unlocks the value of the data, right? It's very easy to use somebody like one of my good, smart friends here, Steven Eliuk to develop models within a domain. But when you talk about cross-functional, complex data coming together to enable models, that's like the Holy Grail of transformation. Then we can deliver real business value. Then you're not waiting to make decisions. Then you can actually be ahead of trends. And so that's what we've been trying to do And the thought and the journey that we have been on is build a enterprise data platform. So, take the concept of a data lake. Bring in all your data sources into one place, but on top of that, make it more than just a data lake. Bring the services and capabilities that allow you to deliver insights from data together with the data so we have a data platform. And our Cognitive Enterprise data platform sort of enables that transformation, and it makes people like my good friend here much more productive and much more valuable to the business. >> This sounds like just a massive challenge. It's not just a technology challenge, obviously. You've got cultural. I mean, people, "This is my data." >> Yes. >> (laughs) And I'm referring, Tim, you're talking like you're largely through this process, right? So it first of all is... Can you talk about-- >> Basically, I will say this. This is a journey. You're never done, right? And one of the reasons why it is a journey is, if you're going to have a successful business, your business is going to keep transforming. Things are going to keep changing. And even in our landscape today, regulations are going to come. So there's always going to be some type of challenge. So I like to say, we're in a journey. We're not finished. (laughing) We're well down the path, and we've learned a lot. And one of the things we have learned, you hit on it, is culture, right? And it's a little hard to say, okay, I'm opening things up. I don't own the data. The company owns the data. There is that sort of cultural change that has to go along with this transformation. >> And there are technology challenges. I mean, when I first started in this business, AI was a hot concept, but you needed, like, massive supercomputers to actually make them work. Today, you now see their sort of rebirth. You know, (mumbling) talks about the AI winter, and now it's like the AI spring. >> Yeah. >> So how are you guys applying machine intelligence to make IBM a better business? >> Well, ultimately, the technology is really, basically transitioned us from the Dark Ages forward. Previously in the supercomputer mentality, didn't fit well for a lot of AI tasks. Now with GPUs and accelerators and FBGAs and things like that, we're definitely able, along with the data and the curated data that we need, to just fast-track. You know, the practitioners would spend an amazing amount of time gathering, crowdsourcing data, getting it in good order, and then the computational challenges were tough. Now, IBM came to the market with a very interesting computer. The POWER8 and POWER9 architecture has NVLink, which is a proprietary Nvidia, interconnect directly to the CPU. So we can feed GPUs a lot quicker for certain types of tasks. And for certain types of tasks that could mean, you know, you get to market quicker, or we get insights for enterprise problems quicker. So technology's a big deal, but it doesn't just center around GPUs. If you're slow to get access to the data, then that's a big problem. So the governance (mumbling) aspects are just as important, in addition to that, security, privacy, et cetera, also important. The quality of the data, where the data is. So it's and end-to-end system, and if there's any sort of impedance on any of it, it slows down the entire process. But then you have very expensive practitioners who are trying to do their job that are waiting on data or waiting on results. So it's really an end-to-end process. >> Okay, so let's assume for a second the technology box is checked. And again, as you say, Tim, it's a journey, and technology's going to continue to evolve. But we're at a point in technology now where this stuff actually can work. But what about data quality? What about compliance and governance? How are you dealing with the natural data quality problem? Because I'm a PNL manager. I'm saying, well, we're making data decisions, but if I don't like the decision, I'm going to attack the quality of the data. (laughing) So who adjudicates all that, and how have you resolved those challenges? >> Well, I like to think of... I'm an engineer by study, and I just like to think of simple formulas. Garbage in, garbage out. It applies to everything, and it definitely applies to data. >> (laughs) >> Your insights, the models, anything that you build is only going to be as good as the data foundation you have. So one of the key things that we've embarked on a journey on is, how do we standardize all aspects of data across the company? Now, you might say, hey, that's not a hard challenge, but it's really easy to do standards in a silo. For this organization, this is how we're going to call terms like geography, and this is how we'll represent these other terms. But when you do that across functions, it becomes conflict, right? Because people want to do it their own way. So we're on the path of standardizing data across the enterprise. That's going to allow us to have good definitions. And then, as you mentioned earlier, we are trying to use AI to be able to improve our data quality. One of the most important things about data is the metadata, the data that describes the data. >> Mm-hm. >> And we're trying to use AI to enhance our metadata. I'd love for Steven to talk a little bit about this, 'cause this is sort of his brainchild. But it's fascinating to me that we can be on a AI transformation, data can be at the heart of it, and we can use AI (laughs) to help improve the quality of our data. >> Right. >> It's fascinating. >> So the metadata problem is (mumbling) because you've talked about data length before. Then in this day and age, you're talking schema lists. Throw it into a data lake and figure out because you have to be agile for your business. So you can't do that with just human categorization, and you know, it's got to-- >> It could take hours, maybe years. >> For a company the size of IBM, the market would shift so fast, right? So how do you deal with that problem? >> That's exactly it. We're not patient enough to do the normative kind of mentality where you just throw a whole bunch of bodies at it. We're definitely moving from that non-extensible man count, full-time-employee type situation, to looking for ways that we can utilize automation. So around the metadata, quality and understanding of that data was incredibly problematic, and we were just hiring people left, right, and center. And then it's a really tough job that they have dealing with so many different business islands, et cetera. So looking for ways that we could automate that process, we finally found away to do it. So there's a lot of curated data. Now we're looking at data quality in addition to looking at regulatory and governance issues, in addition to automating the labeling of business metadata. And the business metadata is the taxonomy that everything is linked together. We understand it under the same normative umbrella. So then when one of the enterprise use cases says, "Hey, we're looking for additional data assets," oh, it's (snaps) in the cloud here, or it's in a private instance here. But we know it's there, and you can grab it, right? So we're definitely at probably the tail end of that curve now, and it started off really hard, but it's getting easier. So that's-- >> Guys, we got to leave it there. Awesome discussion. I hope we can pick it up in the future when maybe we have more metadata than data. >> (laughs) >> And metadata's going to become more and more valuable. But thank you so much for sharing a little bit about IBM's transformation. It was great having you guys on. >> Thank you. >> Alright, keep it right there, everybody. We'll be back with our next guest right after this short break. You're watching the Cube at IBM CDO in San Francisco. Right back. (electronic music) >> Alright, long clear. Alright, thank you guys. Appreciate it, I wish we had more time.

>> From our studios, in the heart of Silicon Valley, Palo Alto, California, this is a CUBE Conversation. >> Welcome to this CUBE Conversation here in Palo Alto, California. I'm John Furrier, host of theCUBE. We are here with Arpit Joshipura, GM of Networking, Edge, IoT for the Linux Foundation. Arpit, great to see you again, welcome back to theCUBE, thanks for joining us. >> Thank you, thank you. Happy to be here. >> So obviously, we love the Linux Foundation. We've been following all the events; we've chatted in the past about networking. Computer storage and networking just doesn't seem to go away with cloud and on-premise hybrid cloud, multicloud, but open-source software continues to surpass expectations, growth, geographies outside the United States and North America, just overall, just greatness in software. Everything's an abstraction layer now; you've got Kubernetes, Cloud Native- so many good things going on with software, so congratulations. >> Well thank you. No, I think we're excited too. >> So you guys got a big event coming up in China: OSS, Open Source Summit, plus KubeCon. >> Yep. >> A lot of exciting things, I want to talk about that in a second. But I want to get your take on a couple key things. Edge and IoT, deep learning and AI, and networking. I want to kind of drill down with you. Tell us what's the updates on the projects around Linux Foundation. >> Okay. >> The exciting ones. I mean, we know Cloud Native CNCF is going to take up more logos, more members, keeps growing. >> Yep. >> Cloud Native clearly has a lot of opportunity. But the classic in the set, certainly, networking and computer storage is still kicking butt. >> Yeah. So, let me start off by Edge. And the fundamental assumption here is that what happened in the cloud and core is going to move to the Edge. And it's going to be 50, 100, 200 times larger in terms of opportunity, applications, spending, et cetera. And so what LF did was we announced a very exciting project called Linux Foundation Edge, as an umbrella, earlier in January. And it was announced with over 60 founding members, right. It's the largest founding member announcement we've had in quite some time. And the reason for that is very simple- the project aims at unifying the fragmented edge in IoT markets. So today, edge is completely fragmented. If you talk to clouds, they have a view of edge. Azure, Amazon, Baidu, Tencent, you name it. If you talk to the enterprise, they have a view of what edge needs to be. If you talk to the telcos, they are bringing the telecom stack close to the edge. And then if you talk to the IoT vendors, they have a perception of edge. So each of them are solving the edge problems differently. What LF Edge is doing, is it is unifying a framework and set of frameworks, that allow you to create a common life cycle management framework for edge computing. >> Yeah. >> Now the best part of it is, it's built on five exciting technologies. So people ask, "You know, why now?" So, there are five technologies that are converging at the same time. 5G, low latency. NFV, network function virtualization, so on demand. AI, so predictive analytics for machine learning. Container and microservices app development, so you can really write apps really fast. And then, hardware development: TPU, GPU, NPU. Lots of exciting different size and shapes. All five converging; put it close to the apps, and you have a whole new market. >> This is, first of all, complicated in the sense of... cluttered, fragmented, shifting grounds, so it's an opportunity. >> It's an opportunity. >> So, I get that- fragmented, you've got the clouds, you've got the enterprises, and you've got the telcos all doing their own thing. >> Yep. >> So, multiple technologies exploding. 5G, Wi-Fi 6, a bunch of other things you laid out, >> Mhmm. >> all happening. But also, you have all those suppliers, right? >> Yes. >> And, so you have different manufacturers-- >> And different layers. >> So it's multiple dimensions to the complexity. >> Correct, correct. >> What are you guys seeing, in terms of, as a solution, what's motivating the founding members; when you say unifying, what specifically does that mean? >> What that means is, the entire ecosystem from those markets are coming together to solve common problems. And I always sort of joke around, but it's true- the common problems are really the plumbing, right? It's the common life cycle management, how do you start, stop, boot, load, log, you know, things like that. How do you abstract? Now in the Edge, you've 400, 500 interfaces that comes into an IoT or an edge device. You know, Zigbee, Bluetooth, you've got protocols like M2T; things that are legacy and new. Then you have connectivity to the clouds. Devices of various forms and shapes. So there's a lot of end by end problems, as we call it. So, the cloud players. So for LF Edge for example, Tencent and Baidu and the cloud leaders are coming together and saying, "Let's solve it once." The industrial IoT player, like Dynamic, OSIsoft, they're coming in saying, "Let's solve it once." The telcos- AT&T, NTT, they're saying "Let's solve it once. And let's solve this problem in open-source. Because we all don't need to do it, and we'll differentiate on top." And then of course, the classic system vendors that support these markets are all joining hands. >> Talk about the business pressure real quick. I know, you look at, say, Alibaba for instance, and the folks you mentioned, Tencent, in China. They're perfecting the edge. You've got videos at the edge; all kinds of edge devices; people. >> Correct. >> So there's business pressures, as well. >> The business pressure is very simple. The innovation has to speed up. The cost has to go down. And new apps are coming up, so extra revenue, right? So because of these five technologies I mentioned, you've got the top killer apps in edge are anything that is, kind of, video but not YouTube. So, anything that the video comes from 360 venues, or drones, things like that. Plus, anything that moves, but that's not a phone. So things like connected cars, vehicles. All of those are edge applications. So in LF Edge, we are defining edge as an application that requires 20 milliseconds or less latency. >> I can't wait for someone to define- software define- "edge". Or, it probably is defined. A great example- I interviewed an R&D engineer at VMware yesterday in San Francisco, it was at the RADIO event- and we were just riffing on 5G, and talking about software at the edge. And one of the advances >> Yes. >> that's coming is splicing the frequency so that you can put software in the radios at the antennas, >> Correct. Yeah. >> so you can essentially provision, in real time. >> Correct, and that's a telco use case, >> Yeah. >> so our projects at the LF Edge are EdgeX Foundry, Akraino, Edge Virtualization Engine, Open Glossary, Home Edge. There's five and growing. And all of these software projects can allow you to put edge blueprints. And blueprints are really reference solutions for smart cities, manufacturing, telcos, industrial gateways, et cetera et cetera. So, lots of-- >> It's kind of your fertile ground for entrepreneurship, too, if you think about it, >> Correct; startups are huge. >> because, just the radio software that splices the radio spectrum is going to potentially maybe enable a service provider market, and towers, right? >> Correct, correct. >> Own my own land, I can own the tower and rent it out, one radio. >> Yep. >> So, business model innovations also an opportunity, >> It's a huge-- >> not just the business pressure to have an edge, but-- >> Correct. So technology, business, and market pressures. All three are colliding. >> Yeah, perfect storm. >> So edge is very exciting for us, and we had some new announcements come out in May, and more exciting news to come out in June, as well. >> And so, going back to Linux Foundation. If I want to learn more. >> LFEdge.org. >> That's kind of the CNCF of edge, if you will, right? Kind of thing. >> Yeah. It's an umbrella with all the projects, and that's equivalent to the CNCF, right. >> Yeah. >> And of course it's a huge group. >> So it's kind of momentum. 64 founding members-- >> Huge momentum. Yeah, now we are at 70 founding members, and growing. >> And how long has it been around? >> The umbrella has been around for about five months; some of the projects have been around for a couple of years, as they incubate. >> Well let us know when the events start kicking in. We'll get theCUBE down there to cover it. >> Absolutely. >> Super exciting. Again, multiple dimensions of innovation. Alright, next topic, one of my favorites, is AI and deep learning. AI's great. If you don't have data you can't really make AI work; deep learning requires data. So this is a data conversation. What's going on in the Linux Foundation around AI and deep learning? >> Yeah. So we have a foundation called LF Deep Learning, as you know. It was launched last year, and since then we have significantly moved it forward by adding more members, and obviously the key here is adding more projects, right. So our goal in the LF Deep Learning Foundation is to bring the community of data scientists, researchers, entrepreneurs, academia, and users to collaborate. And create frameworks and platforms that don't require a PhD to use. >> So a lot of data ingestion, managing data, so not a lot of coding, >> Platforms. >> more data analyst, and/or applications? >> It's more, I would say, platforms for use, right? >> Yeah. >> So frameworks that you can actually use to get business outcomes. So projects include Acumos, which is a machine learning framework and a marketplace which allows you to, sort of, use a lot of use cases that can be commonly put. And this is across all verticals. But I'll give you a telecom example. For example, there is a use case, which is drones inspecting base stations-- >> Yeah. >> And doing analytics for maintenance. That can be fed into a marketplace, used by other operators worldwide. You don't have to repeat that. And you don't need to understand the details of machine learning algorithms. >> Yeah. >> So we are trying to do that. There are projects that have been contributed from Tencent, Baidu, Uber, et cetera. Angel, Elastic Deep Learning, Pyro. >> Yeah. >> It's a huge investment for us. >> And everybody wins when there's contribution, because data's one of those things where if there's available, it just gets smarter. >> Correct. And if you look at deep learning, and machine learning, right. I mean obviously there's the classic definition; I won't go into that. But from our perspective, we look at data and how you can share the data, and so from an LF perspective, we have something called a CDLA license. So, think of an Apache for data. How do you share data? Because it's a big issue. >> Big deal. >> And we have solved that problem. Then you can say, "Hey, there's all these machine learning algorithms," you know, TensorFlow, and others, right. How can you use it? And have plugins to this framework? Then there's the infrastructure. Where do you run these machine learning? Like if you run it on edge, you can run predictive maintenance before a machine breaks down. If you run it in the core, you can do a lot more, right? So we've done that level of integration. >> So you're treating data like code. You can bring data to the table-- >> And then-- >> Apply some licensing best practices like Apache. >> Yes, and then integrate it with the machine learning, deep learning models, and create platforms and frameworks. Whether it's for cloud services, for sharing across clouds, elastic searching-- >> And Amazon does that in terms of they vertically integrate SageMaker, for instance. >> That's exactly right. >> So it's a similar-- >> And this is the open-source version of it. >> Got it- oh, that's awesome. So, how does someone get involved here, obviously developers are going to love this, but-- >> LF Deep Learning is the place to go, under Linux Foundation, similar to LF Edge, and CNCF. >> So it's not just developers. It's also people who have data, who might want to expose it in. >> Data scientists, databases, algorithmists, machine learning, and obviously, a whole bunch of startups. >> A new kind of developer, data developer. >> Right. Exactly. And a lot of verticals, like the security vertical, telecom vertical, enterprise verticals, finance, et cetera. >> You know, I've always said- you and I talked about this before, and I always rant on theCUBE about this- I believe that there's going to be a data development environment where data is code, kind of like what DevOps did with-- >> It's the new currency, yeah. >> It's the new currency. >> Yeah. Alright, so final area I want to chat with you before we get into the OSS China thing: networking. >> Yeah. >> Near and dear to your heart. >> Near and dear to my-- >> Networking's hot now, because if you bring IoT, edge, AI, networking, you've got to move things around-- >> Move things around, (laughs) right, so-- >> And you still need networking. >> So we're in the second year of the LF Networking journey, and we are really excited at the progress that has happened. So, projects like ONAP, OpenDaylight, Tungsten Fabric, OPNFV, FDio, I mean these are now, I wouldn't say household names, but business enterprise names. And if you've seen, pretty much all the telecom providers- almost 70% of the subscribers covered, enabled by the service providers, are now participating. Vendors are completely behind it. So we are moving into a phase which is really the deployment phase. And we are starting to see, not just PoCs [Proofs of Concept], but real deployments happening, some of the major carriers now. Very excited, you know, Dublin, ONAP's Dublin release is coming up, OPNFV just released the Hunter release. Lots of exciting work in Fido, to sort of connect-- >> Yeah. >> multiple projects together. So, we're looking at it, the big news there is the launch of what's called OVP. It's a compliance and verification program that cuts down the deployment time of a VNF by half. >> You know, it's interesting, Stu and I always talk about this- Stu Miniman, CUBE cohost with me- about networking, you know, virtualization came out and it was like, "Oh networking is going to change." It's actually helped networking. >> It helped networking. >> Now you're seeing programmable networks come out, you see Cisco >> And it's helped. >> doing a lot of things, Juniper as well, and you've got containers in Kubernetes right around the corner, so again, this is not going to change the need, it's going to- It's not going to change >> It's just a-- >> the desire and need of networking, it's going to change what networking is. How do you describe that to people? Someone saying, "Yeah, but tell me what's going on in networking? Virtualization, we got through that wave, now I've got the container, Kubernetes, service mesh wave, how does networking change? >> Yeah, so it's a four step process, right? The first step, as you rightly said, virtualization, moved into VMs. Then came disaggregation, which was enabled by the technology SDN, as we all know. Then came orchestration, which was last year. And that was enabled by projects like ONAP and automation. So now, all of the networks are automated, fully running, self healing, feedback closed control, all that stuff. And networks have to be automated before 5G and IoT and all of these things hit, because you're no longer talking about phones. You're talking about things that get connected, right. So that's where we are today. And that journey continues for another two years, and beyond. But very heavy focused on deployment. And while that's happening, we're looking at the hybrid version of VMs and containers running in the network. How do you make that happen? How do you translate one from the other? So, you know, VNFs, CNFs, everything going at the same time in your network. >> You know what's exciting is with the software abstractions emerging, the hard problems are starting to emerge because as it gets more complicated, end by end problems, as you said, there's a lot of new costs and complexities, for instance, the big conversation at the Edge is, you don't want to move data around. >> No, no. >> So you want to move compute to the edge, >> You can, yeah-- >> But it's still a networking problem, you've still got edge, so edge, AI, deep learning, networking all tied together-- >> They're all tied together, right, and this is where Linux Foundation, by developing these projects, in umbrellas, but then allowing working groups to collaborate between these projects, is a very simple governance mechanism we use. So for example, we have edge working groups in Kubernetes that work with LF Edge. We have Hyperledger syncs that work for telecoms. So LFN and Hyperledger, right? Then we have automotive-grade Linux, that have connected cars working on the edge. Massive collaboration. But, that's how it is. >> Yeah, you connect the dots but you don't, kind of, force any kind of semantic, or syntax >> No. >> into what people can build. >> Each project is autonomous, >> Yeah. >> and independent, but related. >> Yeah, it's smart. You guys have a good view, I'm a big fan of what you guys are doing. Okay, let's talk about the Open Source Summit and KubeCon, happening in China, the week of the 24th of June. >> Correct. >> What's going on, there's a lot of stuff going on beyond Cloud Native and Linux, what are some of the hot areas in China that you guys are going to be talking about? I know you're going over. >> Yeah, so, we're really excited to be there, and this is, again, life beyond Linux and Cloud Native; there's a whole dimension of projects there. Everything from the edge, and the excitement of Iot, cloud edge. We have keynotes from Tencent, and VMware, and all the Chinese- China Mobile and others, that are all focusing on the explosive growth of open-source in China, right. >> Yeah, and they have a lot of use cases; they've been very aggressive on mobility, Netdata, >> Very aggressive on mobility, data, right, and they have been a big contributor to open-source. >> Yeah. >> So all of that is going to happen there. A lot of tracks on AI and deep learning, as a lot more algorithms come out of the Tencents and the Baidus and the Alibabas of the world. So we have tracks there. We have huge tracks on networking, because 5G and implementation of ONAP and network automation is all part of the umbrella. So we're looking at a cross-section of projects in Open Source Summit and KubeCon, all integrated in Shanghai. >> And a lot of use cases are developing, certainly on the edge, in China. >> Correct. >> A lot of cross pollination-- >> Cross pollination. >> A lot of fragmentation has been addressed in China, so they've kind of solved some of those problems. >> Yeah, and I think the good news is, as a global community, which is open-source, whether it's Europe, Asia, China, India, Japan, the developers are coming together very nicely, through a common governance which crosses boundaries. >> Yeah. >> And building on use cases that are relevant to their community. >> And what's great about what you guys have done with Linux Foundation is that you're not taking positions on geographies, because let the clouds do that, because clouds have-- >> Clouds have geographies, >> Clouds, yeah they have agents-- >> Edge may have geography, they have regions. >> But software's software. (laughs) >> Software's software, yeah. (laughs) >> Arpit, thanks for coming in. Great insight, loved talking about networking, the deep learning- congratulations- and obviously the IoT Edge is hot, and-- >> Thank you very much, excited to be here. >> Have a good trip to China. Thanks for coming in. >> Thank you, thank you. >> I'm John Furrier here for CUBE Conversation with the Linux Foundation; big event in China, Open Source Summit, and KubeCon in Shanghai, week of June 24th. It's a CUBE Conversation, thanks for watching.

(light music) >> Live from Boston. It's theCUBE. Covering IBM Chief Data Officer Summit. Brought to you by IBM. >> Welcome back everyone, to theCUBE's live coverage of the IBM CDO Summit here in Boston, Massachusetts. I'm your host, Rebecca Night. And I'm joined by my co-host, Paul Gillen. We have two guests for this segment. We have Steven Eliuk, who is the Vice President of Deep Learning Global Chief Data Officer at IBM. And Christopher Bannocks, Group Chief Data Officer at IMG. Thanks so much for coming on theCUBE. >> My pleasure. >> Before we get started, Steve, I know you have some very important CUBE fans that you need-- >> I do. >> To give a shout out to. Please. >> For sure. So I missed them on the last three runs of CUBE, so I'd like to just shout out to Santiago, my son. Five years old. And the shortest one, which is Elana. Miss you guys tons and now you're on the air. (all laughing) >> Excellent. To get that important piece of business out. >> Absolutely. >> So, let's talk about Metadata. What's the problem with Metadata? >> The one problem, or the many (chuckles)? >> (laughing) There are a multitude of problems. >> How long ya got? The problem is, it's everywhere. And there's lots of it. And bringing context to that and understanding it from enterprise-wide perspective is a huge challenge. Just connecting to it finding it, or collecting centrally and then understanding the context and what it means. So, the standardization of it or the lack of standardization of it across the board. >> Yeah, it's incredibly challenging. Just the immense scale of metadata at the same time dealing with metadata as Chris mentioned. Just coming up with your own company's glossary of terms to describe your own data. It's kind of step one in the journey of making your data discoverable and governed. Alright, so it's challenging and it's not well understood and I think we're very early on in these stages of describing our data. >> Yeah. >> But we're getting there. Slowly but surely. >> And perhaps in that context it's not only the fact that it's everywhere but actually we've not created structural solutions in a consistent way across industries to be able to structure it and manage it in an appropriate way. >> So, help people do it better. What are some of the best practices for creating, managing metadata? >> Well you can look at diff, I mean, it's such a broad space you can look at different ones. Let's just take the work we do around describing our data and we do that for for the purposes of regulation. For the purposes of GDPR et cetera et cetera. It's really about discovering and providing context to the data that we have in the organization today. So, in that respect it's creating a catalog and making sure that we have the descriptions and the structures of the data that we manage and use in the organization and to give you perhaps a practical example when you have a data quality problem you need to know how to fix it. So, you store, so you create and structure metadata around well, where does it come from, first of all. So what's the journey it's taken to get to the point where you've identified that there's a problem. But also then, who do we go to to fix it? Where did it go wrong in the chain? And who's responsible for it? Those are very simple examples of the metadata around, the transformations the data might have come through to get to its heading point. The quality metrics associated with it. And then, the owner or the data steward that it has to be routed back to to get fixed. >> Now all of those are metadata elements >> All of those, yeah. >> Right? >> 'Cause we're not really talking about the data. The data might be a debit or a credit. Something very simple like that in banking terms. But actually it's got lots of other attributes associated with it which essentially describe that data. So, what is it? Who owns it? What are the data quality metrics? How do I know whether what it's quality is? >> So where do organizations make mistakes? Do they create too much metadata? Do they create poor, is it poorly labeled? Is it not federated? >> Yes. (all laughing) >> I think it's a mix of all of them. One of the things that you know Chris alluded to and you might of understood is that it's incredibly labor-intensive task. There's a lot of people involved. And when you get a lot of people involved in sadly a quite time-consuming, slightly boring job there's errors and there's problem. And that's data quality, that's GDPR, that's government owned entities, regulatory issues. Likewise, if you can't discover the data 'cause it's labeled wrong, that's potential insight that you've now lost. Because that data's not discoverable to a potential project that's looking for similar types of data. Alright, so, kind of step one is trying to scribe your metadata to the organization. Creating a taxonomy of metadata. And getting everybody on board to label that data whether it be short and long descriptions, having good tools et cetera. >> I mean look, the simple thing is... we struggle as... As a capability in any organization we struggle with these terms, right? Metadata, well ya know, if you're talking to the business they have no idea what you're talking about. You've already confused them the minute you mentioned meta. >> Hashtag. >> Yeah (laughs) >> It's a hashtag. >> That's basically what it is. >> Essentially what it is it's just data about data. It's the descriptive components that tell you what it is you're dealing with. If you just take a simple example from finance; An interest rate on it's own tells you nothing. It could be the interest rate on a savings account. It can the interest rate on a bond. But on its own you have no clue, what you're talking about. A maturity date, or a date in general. You have to provide the context. And that is it's relationships to other data and the contexts that it's in. But also the description of what it is you're looking at. And if that comes from two different systems in an organization, let's say one in Spain and one in France and you just receive a date. You don't know what you're looking at. You have not context of what you're looking at. And simply you have to have that context. So, you have to be able to label it there and then map it to a generic standard that you implement across the organization in order to create that control that you need in order to govern your data. >> Are there standards? I'm sorry Rebecca. >> Yes. >> Are there standards efforts underway industry standard why difference? >> There are open metadata standards that are underway and gaining great deal of traction. There are an internally use that you have to standardize anyway. Irrespective of what's happening across the industry. You don't have the time to wait for external standards to exist in order to make sure you standardize internally. >> Another difficult point is it can be region or country specific. >> Yeah. >> Right, so, it makes it incredibly challenging 'cause every region you might work in you might have to have a own sub-glossary of terms for that specific region. And you might have to control the export of certain data with certain terms between regions and between countries. It gets very very challenging. >> Yeah. And then somehow you have to connect to it all to be able to see what it all is because the usefulness of this is if one system calls exactly the same, maps to let's say date. And it's local definition of that is maturity date. Whereas someone else's map date to birthdate you know you've got a problem. You just know you've got a problem. And exposing the problem is part of the process. Understanding hey that mapping's wrong guys. >> So, where do you begin? If your mission is to transform your organization to be one that is data-centric and the business side is sort of eyes glazing over at the mention of metadata. What kind of communication needs to happen? What kind of teamwork, collaboration? >> So, I mean teamwork and collaboration are absolutely key. The communication takes time. Don't expect one blast of communication to solve the problem. It is going to take education and working with people to actually get 'em to realize the importance of things. And to do that you need to start something. Just the communication of the theory doesn't work. No one can ever connect to it. You have to have people who are working on the data for a reason that is business critical. And you need have them experience the problem to recognize that metadata is important. Until they experience the problem you don't get the right amount of traction. So you have to start small and grow. >> And you can use potentially the whip as well. Governance, the regulatory requirements that's a nice one to push things along. That's often helpful. >> It's helpful, but not necessarily popular. >> No, no. >> So you have to give-- >> Balance. >> We're always struggling with that balance. There's a lot of regulation that drives the need for this. But equally, that same regulation essentially drives all of the same needs that you need for analytics. For good measurement of the data. For growth of customers. For delivering better services to customers. All of these things are important. Just the web click information you have that's all essentially metadata. The way we interact with our clients online and through mobile. That's all metadata. So it's not all whip or stick. There's some real value that is in there as well. >> These would seem to be a domain that is ideal for automation. That through machine learning contextualization machines should be able to figure a lot of this stuff out. Am I wrong? >> No, absolutely right. And I think there's, we're working on proof of concepts to prove that case. And we have IBM AMG as well. The automatic metadata generation capability using machine learning and AI to be able to start to auto-generate some of this insight by using existing catalogs, et cetera et cetera. And we're starting to see real value through that. It's still very early days but I think we're really starting to see that one of the solutions can be machine learning and AI. For sure. >> I think there's various degrees of automation that will come in waves for the next, immediately right now we have certain degrees where we have a very small term set that is very high confidence predictions. But then you want to get specific to the specificity of a company which have 30,000 terms sometimes. Internally, we have 6,000 terms at IBM. And that level of specificity to have complete automation we're not there yet. But it's coming. It's a trial. >> It takes time because the machine is learning. And you have to give the machine enough inputs and gradually take time. Humans are involved as well. It's not about just throwing the machine at something and letting it churn. You have to have that human involvement. It takes time to have the machine continue to learn and grow and give it more terms. And give it more context. But over time I think we're going to see good results. >> I want to ask about that human-in-the-loop as IBM so often calls it. One of the things that Nander Paul Bendery was talking about is how the CDO needs to be a change engine in chief. So how are the rank and file interpreting this move to automation and increase in machine learning in their organizations? Is it accepted? It is (chuckles) it is a source of paranoia and worry? >> I think it's a mix. I think we're kind of blessed at least in the CDO at IBM, the global CDO. Is that everyone's kind of on board for that mission. That's what we're doing >> Right, right. >> There's team members 25, 30 years on IMBs roster and they're just as excited as I am and I've only been there for 16 months. But it kind of depends on the project too. Ones that have a high impact. Everyone's really gung ho because we've seen process times go from 90 days down to a couple of days. That's a huge reduction. And that's the governance regulatory aspects but more for us it's a little bit about we're looking for the linkage and availability of data. So that we can get more insights from that data and better outcomes for different types of enterprise use cases. >> And a more satisfying work day. >> Yeah it's fun. >> That's a key point. Much better to be involved in this than doing the job itself. The job of tagging and creating metadata associated with the vast number of data elements is very hard work. >> Yeah. >> It's very difficult. And it's much better to be working with machine learning to do it and dealing with the outliers or the exceptions than it is chugging through. Realistically it just doesn't scale. You can't do this across 30,000 elements in any meaningful way or a way that really makes sense from a financial perspective. So you really do need to be able to scale this quickly and machine learning is the way to do it. >> Have you found a way to make data governance fun? Can you gamify it? >> Are you suggesting that data governance isn't fun? (all laughing) Yes. >> But can you gamify it? Can you compete? >> We're using gamification in various in many ways. We haven't been using it in terms of data governance yet. Governance is just a horrible word, right? People have really negative connotations associated with it. But actually if you just step one degree away we're talking about quality. Quality means better decisions. And that's actually all governance is. Governance is knowing where your data is. Knowing who's responsible for fixing if it goes wrong. And being able to measure whether it's right or wrong in the first place. And it being better means we make better decisions. Our customers have better engagement with us. We please our customers more and therefore they hopefully engage with us more and buy more services. I think we should that your governance is something we invented through the need for regulation. And the need for control. And from that background. But realistically it's just, we should be proud about the data that we use in the organization. And we should want the best results from it. And it's not about governance. It's about us being proud about what we do. >> Yeah, a great note to end on. Thank you so much Christopher and Steven. >> Thank you. >> Cheers. >> I'm Rebecca Night for Paul Gillen we will have more from the IBM CDO Summit here in Boston coming up just after this. (electronic music)

>> Announcer: Live from Las Vegas, it's theCUBE, covering NetApp Insight 2018. Brought to you by NetApp. >> Welcome back to theCUBE, we are live. We've been here all day at NetApp Insight in Las Vegas at the Mandalay Bay. I'm Lisa Martin with Stu Miniman and we're joined by a couple of guests. One of our alumni, Bharat Badrinath, the V.P. of Product Solutions and Marketing at NetApp. Hey, Bharat, welcome back. >> Thank you, thanks for having me. >> And we've also got Renee Yao, who is a Senior Product Marketing Manager for Deep Learning and AI Systems at Nvidia. Renee, welcome to theCUBE. >> Thanks for having me. >> So guys, this is a pretty big event. NetApp's biggest customer-partner event, the keynote, standing room only this morning five thousand plus people, lot of buzz, lot of momentum. Speaking of momentum, NetApp and Nvidia just launched an interesting partnership a couple months ago. Bharat, talk to us about how NetApp is working with Nvidia to really take advantage of AI and allow your customers to do that as well. >> Sure. So, as we started talking to customers and started looking at what they were investing in, AI bubbled up, right up to the top. And given our rich history in NFS, high performance NFS, it became an obvious choice for NetApp to invest in this space. So we've been working with Nvidia for a really long time, probably close to a year, to start integrating our products with their DGX-1 supercomputer and providing it as a single package to our customers, which makes it a lot easier for them to deploy their AI instead of waiting months for testing infrastructure, which the data scientists don't want to do. We get them a pre-tested, pre-validated system and our All-Flash Fast, which has been winning multiple awards and the recent A800 announcement were perfect choice for us to integrate into this architecture for the system. >> Alright, Renee, in the keynote this morning, the Futurist, he said-- We talked about data as the new oil, he said AI is the new electricity. Maybe you can speak a little bit as to why this is so important. Having gone to a lot of shows this year, it felt like every single show I go to, I see Nvidia, arm in arm with partners, because there's a huge wave coming. >> Yes, absolutely, and I think there was this hype about data, there was this hype about AI, and I think the years of Big Data World, that's creating data, absolutely the foundation for AI, and AI as the new electricity is a very, very good analogy. And let's do some math, shall we? So Swiss Federal Railway, it's a very good customer of ours. For those of you who don't know, they're kind of like the heart or center of all the railway tracks going through, serving about 1.2 million passengers on a day-to-day basis. Securing their security is very, very important. Now, they also have a lot of switches that turn on, then the train can go by and with the tunnels and bridges and switches, so they need to make sure that these trains actually don't collide. So when one train goes by with 11 switches, that gives you 30 ways of possible routing. Two trains, 900 ways. 80 trains, 10 to the eightieth power of ways. That's more than the observed atoms in the universe. And they actually have more than 10 thousand trains. So think about, can human being possibly calculate that much data and possibilities in their brain? As smart as we all want to think we all are, they turn to DGX, and the full day of simulation on DGX-1 was only 17 seconds for them to get back results. And I think that analogy of AI as the new electricity, just talking about the speed of light, is very spot on. >> So this isn't hype anymore, this is actually reality. And you gave a really great example of how a large transportation system is using it to get almost real time information. Bharat, talk to us about NetApp storage, history, 26 years, you guys have really made a lot of pivots in terms of your digital transformation, your cultural transformation. How are you helping with, now, kind of the added power of Nvidia, helping customers to, the hype's gone, actually deploy it, live it, and benefit a business from it? >> Yeah, absolutely, I think, as you rightly pointed out, NetApp has made a lot of pivots. Right, and I think the latest journey in terms of being empowering our customers with data has been a very powerful mission for the company. We entered the Flash market a little bit later than our competitors, but we have made dramatic progress in that space. In fact, recently, based on the latest IDC report, we were number one in All-Flash market worldwide, so that is quite an accomplishment for a company which was late to the market. And having said that, that's because of the innovation engine that is still alive and well within NetApp. We're announcing, as you've seen in the conference, we're announcing a lot of new products and technology which are way ahead of what our competitors are offering, but I think it is all hinged on what our customers need. The customer benefits because, yeah, it has profound benefit of changing how customers operate, their entire operations, it can transform dramatically overnight. And as Renee pointed out, Big Data gave the foundation which collected all the data, but wasn't able to process it. But AI with the power of Nvidia and DGX is able to utilize that to create those outcomes for customers. And from our perspective, we bring two key value adds to the space. One, we're able to serve up the data at incredibly high speeds with our award-winning All-Flash systems. But more importantly, data today lives everywhere. If you think about it, edge is becoming even more important. You can't expect an autonomous car to make an instantaneous decision without the backing of data, which means it can't, everything can't reside in the cloud, it may be at the edge. Some of it may be at your data center. How do you tie all three together, edge, core, and cloud? And that's where the data fabric, the vision of data fabric that you saw today comes in the picture. So one is performance, the ability to stream up the kind of data at the speed of the new processors are demanding, at the speed the customers are demanding to make business decisions and also the edge to core to cloud, our data fabric, which is unique and unparalleled in the industry. >> Now, I'm wondering if you could both bring us inside the customers a little bit. If I think of the traditional storage customer, I need performance, I have more and more data that I need to deal with. But Renee pointed out real outcomes, which is beyond what a traditional storage person would be doing. Who are you working with at the customers-- How do they put together-- It almost sounds like you're building a car. I've got the engine, I've got all the pieces. Who helps put this whole solution together? How does the partnership on the customer's side go together? >> That's a great question. I'll give my take and you can jump on it because she's just returned from being on road shows with joint customers and prospects. So I believe it has to be a joint decision. It's not like IT does it first and the data scientists come in later. Although it may be the case in certain instances where the data scientists start the discussion and then the IT gets brought in. In an ideal case, just like building a car, you want all the teams to be sitting together, make sure they're making the right calls because every compromise you make at one end will impact the other. So you want to make sure you make the optimal decision end to end. And that's where some of our channel partners come in who kind of bridge the data scientist team and the IT team. In some cases, customers show up with data scientists and IT teams together and some, it's one after the other. >> Absolutely. We see the same thing when we're on the road show. Literally two weeks ago, in Canada, by the way, there was a snowstorm, and it was an unforeseen snowstorm, you don't get snowstorm in October-- >> Yes, even for Canada, it was unforeseen. >> Yeah, and we had a packed room of people coming to learn about AI and in the audience, we absolutely see people from the infrastructure side, from the data center side, from the data scientist side, and they realized that they really have to start talking because none of them can afford to be reactive. For example, the data scientists, we want to do the innovation. I can't just go to the infrastructure guys and say that, "Hey, this is my workload, do something about it." And the infrastructure guys don't want to hold on to that problem and then don't know what to do with it. They really need to be ahead of everything and I think the interesting thing is, among those four cities that we're at, we see customers from the government, oil and gas, transportation, health care, and just any industry you can think of, they're all here. One specific example, do you know Mike's company that actually came to us, they have about 15 petabytes of data and that's storing 20 years of historical data and they only have two staff and they were not hiring more staff. They were like, "We just want something that's "going to be able to work and we know everything, "so just give us a solution that's going to be able to "easily scale up and out and enable us to continue to "store more data, manage more data, "and get insights out of data fast." So they came to both of us, it's just a very good, natural decision. That's why we have a partnership together as well. >> So you guys talked about kind of connecting the data scientists with the infrastructure folks. Where's the business involved in this conversation? In terms of, we want to identify new products and services to deliver faster than our competition, new markets. Talk to us about, are the data scientists and the infrastructure guys and girls following business initiatives that have been set or are the business leaders involved in these joint conversations? >> Go ahead, you take it. >> Sure. So, I think we see both. We definitely see that there's top-level executives saying that this is our initiative and we have to do it. And they will make the decision that we have to refresh our infrastructure from the ground up to make sure we're supportive of our data scientists' innovation. We've also seen brilliant minds, researchers, data scientists doing amazing things and then roll it up to the VP level and then roll it up to CEO level to say that this has to be done because this-- For example, that simulation of 17 second results, it's things that people used to cannot do in their lifetime, now they can do it in seconds, that kind of innovation just cannot be ignored. >> Yeah, we see the same thing. In fact, any team that has possession of that data or is accountable for that data is the one usually driving the decisions. Because as you mine the data, as you start deploying new techniques, you realize new opportunities, which means the business gets more interested in it and vice versa. If the business is interested, they're going to look for those answers within the data that they have. >> So last thing, Renee, you were on the Women in Tech panel that ended yesterday, Bharat and I were both in the audience, and one of the things that I thought was really inspiring about your story is that you had given us, the audience, an interesting example of a TV opportunity that you were inspired to do by the CEO of Nvidia. Give our audience who didn't have a chance to see that panel a little bit, and in the last minute, of that story and how you were able to step forward and go, "I'm going to try this." >> Yeah, of course. I think that brings us back to the concept that we have at Nvidia, the speed of light concept, and you really have to learn, act, to move at the speed of light, just like our GPUs, with extreme performance. And obviously, at that speed, none of us know everything. So what Jensen, CEO, shared with us was, in an all-hands meeting internally, he told us that none of us are here qualified to do any of our jobs, maybe besides his legal counsel and CFO. And all of us are here to learn, and we need to learn as fast and as much as we can. And we can't really just let the competition determine where our limit is, but instead is by the limit of what is possible. So that is very much a fundamental mindset change in this AI revolution. >> Well thanks so much, Renee and Bharat, for stopping by and sharing with us the exciting things that you guys are doing with NetApp. We look forward to talking with you again soon. >> Thank you. >> Me too, thanks. >> For Stu Miniman, I'm Lisa Martin. You're watching theCUBE, live from NetApp Insight 2018 in Las Vegas. Stu and I will be right back with our next guests after a short break. (techno music)

live from New York it's the cube covering AWS summit New York 2018 hot GUI Amazon Web Services and its ecosystem partners hello and welcome back here live cube coverage in New York City for AWS Amazon Web Services summit 2018 I'm John Fourier with Jeff Rick here at the cube our next guest is dr. Matt wood general manager of artificial intelligence with Amazon Web Services keep alumnae been so busy for the past year and been on the cubanía thanks for coming back appreciate you spending the time so promotions keep on going on you got now general manager of the AI group AI operations ai automation machine learning offices a lot of big category of new things developing and a you guys have really taken AI and machine learning to a whole new level it's one of the key value propositions that you guys now have for not just a large enterprise but down to startups and developers so you know congratulations and what's the update oh well the update is this morning in the keynote I was lucky enough to introduce some new capabilities across our platform when it comes to machine learning our mission is that we want to be able to take machine learning and make it available to all developers we joke internally that we just want to we want to make machine learning boring we wanted to make it vanilla it's just it's another tool in the tool chest of any developer and any any data data scientist and we've done that this idea of taking technology that is traditionally only within reached a very very small number of well-funded organizations and making it as broadly distributed as possible we've done that pretty successfully with compute storage and databases and analytics and data warehousing and we want to do the exact same thing for the machine learning and to do that we have to kind of build an entirely new stack and we think of that stack in in three different tiers the bottom tier really for academics and researchers and data scientists we provide a wide range of frameworks open source programming libraries the developers and data scientists use to build neural networks and intelligence they're things like tend to flow and Apache mx9 and by torch and they're really they're very technical you can build you know arbitrarily sophisticated says most she open source to write mostly open source that's right we contribute a lot of our work back to MX net but we also contribute to buy torch and to tend to flow and there's big healthy open source projects growing up around you know all these popular frameworks plus more like chaos and gluon and horror boredom so that's a very very it's a key area for for researchers and academics the next level up we have machine learning platforms this is for developers and data scientists who have data they see in the clout although they want to move to the cloud quickly but they want to be able to use for modeling they want to be able to use it to build custom machine learning models and so here we try and remove as much of the undifferentiated heavy lifting associated with doing that as possible and this is really where sage maker fits in Cersei's maker allows developers to quickly fill train optimize and host their machine learning models and then at the top tier we have a set of AI services which are for application developers that don't want to get into the weeds they just want to get up and running really really quickly and so today we announced four new services really across those their middle tier in that top tier so for Sage maker we're very pleased to introduce a new streaming data protocol which allows you to take data straight from s3 and pump it straight into your algorithm and straight onto the computer infrastructure and what that means is you no longer have to copy data from s3 onto your computer infrastructure in order to be able to start training you just take away that step and just stream it right on there and it's an approach that we use inside sage maker for a lot of our built-in algorithms and it significantly increases the the speed of the algorithm and significantly of course decreases the cost of running the training because you pay by the second so any second you can save off it's a coffin for the customer and they also it helps the machine learn more that's right yeah you can put more data through it absolutely so you're no longer constrained by the amount of disk space you're not even constrained by the amount of memory on the instance you can just pump terabyte after terabyte after terabyte and we actually had another thing like talked about in the keynote this morning a new customer of ours snap who are routinely training on over 100 terabytes of image data using sage maker so you know the ability to be able to pump in lots of data is one of the keys to building successful machine learning applications so we brought that capability to everybody that's using tensorflow now you can just have your tensor flow model bring it to Sage maker do a little bit of wiring click a button and you were just start streaming your data to your tents upload what's the impact of the developer time speed I think it is it is the ability to be able to pump more data it is the decrease in time it takes to start the training but most importantly it decreases the training time all up so you'll see between a 10 and 25 percent decrease in training time some ways you can train more models or you can train more models per in the same unit time or you can just decrease the cost so it's a completely different way of thinking about how to train over large amounts of data we were doing it internally and now we're making it available for everybody through tej matrix that's the first thing the second thing that we're adding is the ability to be able to batch process and stage make them so stage maker used to be great at real-time predictions but there's a lot of use cases where you don't want to just make a one-off prediction you want to predict hundreds or thousands or even millions of things all at once so let's say you've got all of your sales information at the end of the month you want to use that to make a forecast for the next month you don't need to do that in real-time you need to do it once and then place the order and so we added batch transforms to Sage maker so you can pull in all of that data large amounts of data batch process it within a fully automated environment and then spin down the infrastructure and you're done it's a very very simple API anyone that uses a lambda function it's can take advantage of this again just dramatically decreasing the overhead and making it so much easier for everybody to take advantage of machine load and then at the top layer we had new capabilities for our AI services so we announced 12 new language pairs for our translation service and we announced new transcription so capability which allows us to take multi-channel audio such as might be recorded here but more commonly on contact centers just like you have a left channel on the right channel for stereo context centers often record the agent and the customer on the same track and today you can now pass that through our transcribed service long-form speech will split it up into the channels or automatically transcribe it will analyze all the timestamps and create just a single script and from there you can see what was being talked about you can check the topics automatically using comprehend or you can check the compliance did the agents say the words that they have to say for compliance reasons at some point during the conversation that's a material new capability for what's the top surface is being used obviously comprehend transcribe and barri of others you guys have put a lot of stuff out there all kinds of stuff what's the top sellers top use usage as a proxy for uptake you know I think I think we see a ton of we see a ton of adoption across all of these areas but where a lot of the momentum is growing right now is sage maker so if you look at a formula one they just chose Formula One racing they just chose AWS and sage maker as their machine learning platform the National Football League Major League Baseball today announcer they're you know re offering their relationship and their strategic partnership with AWS cream machine learning so all of these groups are using the data which just streams out of these these races all these games yeah and that can be the video or it can be the telemetry of the cars or the telemetry of the players and they're pumping that through Sage maker to drive more engaging experiences for their viewers so guys ok streaming this data is key this is a stage maker quickly this can do video yeah just get it all in all of it well you know we'd love data I would love to follow up on that so the question is is that when will sage maker overtake Aurora as the fastest growing product in history of Amazon because I predicted that reinvent that sage maker would go on err is it looking good right now I mean I sorta still on paper you guys are seeing is growing but see no eager give us an indicator well I mean I don't women breakout revenue per service but even the same excitement I'll say this the same excitement that I see Perseids maker now and the same opportunity and the same momentum it really really reminds me of AWS ten years ago it's the same sort of transformative democratizing approach to which really engages builders and I see the same level of the excitement as levels are super super high as well no super high in general reader pipe out there but I see the same level of enthusiasm and movement and the middle are building with it basically absolutely so what's this toy you have here I know we don't have a lot of time but this isn't you've got a little problem this is the world's first deep learning in April were on wireless video camera we thought it D blends we announced it and launched it at reinvent 2017 and actually hold that but they can hold it up to the camera it's a cute little device we modeled it after wall-e the Pixar movie and it is a HD video camera on the front here and in the base here we have a incredibly powerful custom piece of machine learning hardware so this can process over a billion machine learning operations per second you can take the video in real time you send it to the GPU on board and we'll just start processing the stream in real time so that's kind of interesting but the real value of this and why we designed it was we wanted to try and find a way for developers to get literally hands-on with machine learning so the way that build is a lifelong learners right they they love to learn they have an insatiable appetite for new information and new technologies and the way that they learn that is they experiment they start working and they kind of spin this flywheel where you try something out it works you fiddle with it it stops working you learn a little bit more and you want to go around around around that's been tried and tested for developers for four decades the challenge with machine learning is doing that is still very very difficult you need a label data you need to understand the algorithms it's just it's hard to do but with deep lens you can get up and running in ten minutes so it's connected back to the cloud it's good at about two stage makeup you can deploy a pre-built model down onto the device in ten minutes to do object detection we do some wacky visual effects with neural style transfer we do hot dog and no hot dog detection of course but the real value comes in that you can take any of those models tear them apart so sage maker start fiddling around with them and then immediately deploy them back down onto the camera and every developer on their desk has things that they can detect there are pens and cups and people whatever it is so they can very very quickly spin this flywheel where they're experimenting changing succeeding failing and just going round around a row that's for developers your target audience yes right okay and what are some of the things that have come out of it have you seen any cool yes evolutionary it has been incredibly gratifying and really humbling to see developers that have no machine learning experience take this out of the box and build some really wonderful projects one in really good example is exercise detection so you know when you're doing a workout they build a model which detects the exerciser there and then detects the reps of the weights that you're lifting now we saw skeletal mapping so you could map a person in 3d space using a simple camera we saw security features where you could put this on your door and then it would send you a text message if it didn't recognize who was in front of the door we saw one which was amazing which would read books aloud to kids so you would hold up the book and they would detect the text extract the text send the text to paly and then speak aloud for the kids so there's games as educational tools as little security gizmos one group even trained a dog detection model which detected individual species plug this into an enormous power pack and took it to the local dog park so they could test it out so it's all of this from from a cold start with know machine learning experience you having fun yes absolutely one of the great things about machine learning is you don't just get to work in one area you get to work in you get to work in Formula One and sports and you get to work in healthcare and you get to work in retail and and develop a tool in CTO is gonna love this chief toy officers chief toy officers I love it so I got to ask you so what's new in your world GM of AI audition intelligence what does that mean just quickly explain it for our our audience is that all the software I mean what specifically are you overseeing what's your purview within the realm of AWS yeah that's that's a totally fair question so my purview is I run the products for deep learning machine learning and artificial intelligence really across the AWS machine learning team so I get I have a lot of fingers in a lot of pies I get involved in the new products we're gonna go build out I get involved in helping grow usage of existing products I get it to do a lot of invention it spent a ton of time with customers but overall work with the rest of the team on setting the technical and pronto strategy for machine learning at AWS when what's your top priorities this year adoption uptake new product introductions and you guys don't stop it well we do sync we don't need to keep on introducing more and more things any high ground that you want to take what's what's the vision I didn't the vision is to is genuinely to continue to make it as easy as possible for developers to use Ruggiero my icon overstate the importance or the challenge so we're not at the point where you can just pull down some Python code and figure it out we're not even we don't have a JVM for machine learning where there's no there's no developer tools or debuggers there's very few visualizers so it's still very hard if you kind of think of it in computing terms we're still working in assembly language and you're seen learning so there's this wealth of opportunity ahead of us and the responsibility that I feel very strongly is to be able to continually in crew on the staff to continually bring new capabilities to mortar but well cloud has been disrupting IT operations AI ops with a calling in Silicon Valley and the venture circuit Auto ml as a term has been kicked around Auto automatic machine learning you got to train the machines with something data seems to be it strikes me about this compared to storage or compared to compute or compared to some of the core Amazon foundational products those are just better ways to do something they already existed this is not a better way to do something that are exists this is a way to get the democratization at the start of the process of the application of machine learning and artificial intelligence to a plethora of applications in these cases that is fundamentally yeah different in it just a step up in terms of totally agree the power to the hands of the people it's something which is very far as an area which is very fast moving and very fast growing but what's funny is it totally builds on top of the cloud and you really can't do machine learning in any meaningful production way unless you have a way that is cheap and easy to collect large amounts of data in a way which allows you to pull down high-performance computation at any scale that you need it and so through the cloud we've actually laid the foundations for machine learning going forwards and other things too coming oh yes that's a search as you guys announced the cloud highlights the power yet that it brings to these new capabilities solutely yeah and we get to build on them at AWS and at Amazon just like our customers do so osage make the runs on ec2 we wouldn't we won't be able to do sage maker without ec2 and you know in the fullness of time we see that you know the usage of machine learning could be as big if not bigger than the whole of the rest of AWS combined that's our aspiration dr. Matt would I wish we had more time to Chad loved shopping with you I'd love to do a whole nother segment on what you're doing with customers I know you guys are great customer focus as Andy always mentions when on the cube you guys listen to customers want to hear that maybe a reinvent will circle back sounds good congratulations on your success great to see you he showed it thanks off dr. Matt would here in the cube was dreaming all this data out to the Amazon Cloud is whether they be hosts all of our stuff of course it's the cube bringing you live action here in New York City for cube coverage of AWS summit 2018 in Manhattan we'll be back with more after this short break

>> From Orlando, Florida it's theCUBE! Covering SAP SAPPHIRE NOW 2018, brought to you by NetApp. >> Welcome to theCUBE I'm Lisa Martin with Keith Townsend and we are in Orlando at SAP SAPPHIRE NOW 2018, where we're in the NetApp booth and talking with lots of partners and we're excited to welcome back to theCUBE, distinguished alumni Jim McHugh from NVIDIA, you are the VP and GM of Deep Learnings and "other stuff" as you said in the keynote. (all laugh) >> Yeah, and other stuff. That's a lot of responsibility! That other stuff, that, you know, that can really pile up! >> That can kill ya. Yeah, exactly. >> So here we are at SAPPHIRE you've been working with SAP in various forms for a long time, this event is enormous, lots of momentum at NVIDIA, what is NVIDIA doing with SAP? >> We're really helping SAP figure out and drive the development of their SAP Leonardo machine learning services so, machine learning, as we saw in the keynote today, with Haaso as a key component of it, and really what it's doing is it's automating a lot of the standard processes that people did, in the interactions, so whether it's closing your invoices at the end of the quarter, and that can take weeks to go through it manually, you can actually do machine learning and deep learning and do that instantaneously, so you can get a continuous close. Things like service ticketing, so when a service ticket comes in, you know, we all know, you pick up the phone, you call 'em and they collect your information, and then they pass you on to someone else that wants to confirm the information, all that can be handled just in a email, because now I know a lot about you when you send me an email I know who you are, know what company you're with, I know your problem 'cause you stated it, and I can route it, using machine learning, to the appropriate person. I can not only route it to the appropriate person I can look up in a knowledge database and say hey, have we seen this answer a question before feed that to the customer service representative, and when they start interacting with the customer they already have a lot of information about them and it's already well underway. >> So from a practical technology perspective we hear a lot about AI, machine learning, NVIDIA obviously leading the way with GPUs and enabling development frameworks to take advantage of machine learning and that compute power. But the enterprise, we'll at that and we're like you know that, we see obvious value, but I need a data scientist, I need a programmer, I need all this capability, from a technical staff perspective, to take advantage of it. How is NVIDIA, SAP, making that easier to consume? >> So most enterprises, if you're just jumpin' in and tryin' to figure it out, you would need all these people, you'd need a data scientist and someone to go through the process. 'Cause AIs, it's a new way of writing software, and you're using data to train the software, so we don't have, we don't put programmers in a room anymore and let 'em code for nine months and out pops software, you know, eventually. We give 'em more and more data, and the data scientist is training it. Well the good news is we're working with SAP and they have the data scientists, they know how SAP apps work, they know how the integration works, they know the workflows of their customers, so they're building the models and then making it available as a service, right? So when you go to the SAP cloud, you're saying I wanna actually take advantage of the SAP service for service ticketing or, you know, I wanna figure out how I can do my invoice processing better, or I'm an HR representative, and I don't wanna spend 60% of my time reading resumes, I wanna actually have an AI do it for me, and then it's a service that you can consume. There, that we do make it possible, like if you have a developer in your enterprise and you say you know what, I'm a big SAP user but I actually wanna develop a custom app or other some things I might do, then SAP makes available the Leonardo machine learning foundation and you can take advantage of that and develop a custom app. And if you have a really big problem and you wanna take it off, NVIDIA's happy to work with you directly and figure out how to solve different problems. And most of our customers are in all three of those, Right? They're consuming the services 'cause they automate things today, they're figuring out, what are the custom apps they need to build around SAP and then they're, you know, they're figuring out some of the product building products or something else that's a much bigger machine learning, deep learning problem. >> So yesterday during Bill McDermott's keynote he talked about tech for good, now there's been a lot of news recently of tech for not-so-good and data privacy, GDPR, you know, compliance going into affect last week, NVIDIA really has been an integral part of this AI renaissance, you talked about, you know, you can help loads of different customers there's so much potential with AI, as Bill McDermott said yesterday, AI to augment humanity. I can imagine, you know, life and death situations like in healthcare, can you give us an example of what you guys are doing with SAP that, you know, maybe is transforming healthcare at a particular hospital? >> Yeah, so one of the great examples I was just talking about is, what Massachusetts General is doing. Massachusetts General is one of the largest research hospitals in the United States, and they're doing a lot of work in AI, to really automate processes that, you know, when you would take your child in to figure out the bone density scan, which basically tells you the bone age of your child, and they compare it to your biological age, and that can tell you a lot of things, is it just a, you know, a growth problem, or is there something more serious to be concerned about. Well, they would do these MRIs, and then you would have to wait for days while the, the technician and the doctor would flip through a textbook from the 1950's, to determine it. Well Massachusetts General automated all that where they actually trained a neural network on all these different scans and all these different components and now you find out in minutes. So it greatly reduces the stress, right? And there's plenty of other project going on and you can see it in determination if that's a cancer cell, or, you know, so many different aspects of it, your retina happens to be an incredible venue into whether you have hypertension, whether you have Malaria, Dengue fever, so things like, you know what, maybe you shouldn't be around anywhere where you're gonna get bit by a mosquito and it's gonna pass it to your family, all that can now be handled, and you don't need expensive healthcare, you can actually take it to a clinician out in the field. So, we love all that. But if you think about the world of SAP which is the, you know, controls the data records of most companies, right? Their supply chain information, their resource information about, you know, what they have available, all that's being automated. So if we think from the production of food where we're having tractors now that they have the ability to go over a plant and say you know what, that needs insecticide or that needs weeds to be removed 'cause it's just bad for the whole component, or that's a diseased plant and I'm gonna remove it, or it just needs water so it can grow, right? That is increasing the production of food in an organic way, then we improve the distribution centers so it doesn't sit as long, right, so that we can actually have drones flying through the warehouses and knowing what needs to be moved first, go from there, we're moving to autonomous driving vehicles and, where deliveries can happen at night when there's not so much traffic, and then we can get the food as fresh as possible and deliver it. So if you think that whole distribution center and just being in the pipeline as being automated, it's doing an incredible amount of good. And then, jumping into the world of autonomous driving vehicles, it's a 10 trillion dollar business that's being changed, radically. >> So as we think about these super complex systems that we're trying to improve, we start to break them down into small components, smaller components, you end up with these scenarios, these edge scenarios, use cases where, you know, whether it's data frequency, data value, or data latency, we have to push to compute out to the edge. Can you talk about use cases where NVIDIA has pushed the technology far out to the edge to take in massive amounts of data, that effectively can't be sent back to the core or to the data center for processing, what are some of these use cases solutions? >> So it's, the world of IOT is changing as well, right, the compute power has to be where it's needed, right, and in any form, so whether that's cloud based, data center based, or at the edge and we have a great customer that is actually doing inspection, oil refineries, bridges, you know, where they spot a crack or some sort of mark where they have to go look at it, well traditionally what you do is you send out a whole team and they build up scaffolding, or they have people repel down to try to inspect it. Well now what we're doing is flying drones and sending wall crawlers up. So they find something, they get data, and then, instead of actually, like you said, putting it, you know, on a truck and taking it back to your data center or trying to figure out how to have enough bandwidth to get there, they're taking one of our products, which is a DGX station, it's basically the equivalent of a half a row of servers, but it's in a single box, water cooled, and they're putting it in vans sitting out in remote areas of Alaska, and retraining the model there on site. So, they get the latest model, they get more intelligence and they just collect it, and they can resend the drones up and then discover more about it. So it really, really is saving, and that saves a lot of money, so you have a group of really smart you know, technicians and people who understand it and a guy who can do the neural network capability instead of a whole team coming up and setting up scaffolding that would cost millions of dollars. >> That reminds me of that commercial that they showed yesterday during general session SAP commercial with Clive Owen the actor, talking about, you mentioned, you know, cracks in oil wells and things like that it just reminded me of that, and what they talked about in that video was really how invisible software, like SAP, is transforming industries, saving lives, I think I saw on their website an example of how they're leveraging AI and technology to reduce water scarcity in India or save the rhino conservation and what you just described with NVIDIA seems to be quite in alignment with the direction that SAP is going. >> Oh absolutely, yeah, I mean we believe in SAP's view of the intelligent enterprise and people gotta remember, enterprise isn't just like the corporate office whatever, enterprises are many different things, alright. Public safety, if you can think about that, that's a big thing we focus on. A really amazing thing that's going on, thinking about using drones for first responders they actually can know what's going on at the scene and when the other people are showing up they know what kind of area they're going into. Or for search and rescue, drones can cover a lot of territory and detect a human faster than a human can, right? And if you can actually find someone within the first 24 hours, chance of survival is so much higher. All of that is, you know, leveraging the exact same technology that we do for looking at our business processes, right, and it's not as, you know, dramatic, it's not gonna show up on the evening news, but honestly, streamlining our business processes, making it happen so much faster and more efficient makes businesses more efficient, you know, it's better for the company, it's better for the employees as well. >> So let's talk about, something that's, that's taboo, financial services, making money with data, or with analytics or machine learning from data, again we have to, John Furrier is here, and we have someone from NVIDIA here, and if we don't bring up blockchain in some type of way he's gonna throw something at his team, so, >> Let's give a shout out to John Furrier. (laughing) >> Give a shout out to John. But from a practical sense, let's subtract the digital currency part of machine, of blockchain, do you see applications for blockchain from a machine learning perspective? >> Yeah, I mean well, if you just boil blockchain down or for trusted networks, right? And you know you heard Bill McDermott say that on stage he called his marketplaces, or areas that he could do for an exchange, it makes total sense. If I can have a trusted way of doing things where I have a common ledger between companies and we know that it's valid, that we can each interchange with, yeah it makes complete sense, right, now we gotta get to the practical imitation of that and we have to build the trust of the companies to understand, okay this technology can take you there, and that's where I think, you know, where we come in with our technology capabilities, ensuring to people that it's reliable and work, SAP comes in with the customer relationships and trusted in what they've been doing in helping people run their business for years, and then it becomes cultural. Like all things, we can kid ourselves in technology that we'll just solve everything, it's a cultural change. I'm gonna share that common ledger, I'm gonna share that common network and feel confident in it, it's something that people have to do and, you know, my take on that always is when the accuracy is so much better, when the efficiency is so much better, when the return is so much better, we get a lot more comfortable. People used to be nervous about giving the grocery store their phone number, right, 'cause they would track their food, right? And today we're just like okay yeah here's my phone number. (Keith laughing) >> So. (laughs) >> Give you a 30 cent discount, here's my number. >> Exactly. We're so cheap. (laughing) >> So we're in the NetApp booth and you guys recently announced a reference, combined reference, AI reference architecture with NetApp, tell us a little bit more about that. >> Yeah, well the little secret behind all the things we just talked about, there's an incredible amount of data, right, and as you collect this data it's really important to store it in a way that it's accessible when you need it. And when you're doing trainings, I have a product that's called DGX-1, DGX-1 takes an incredible amount of data that helps us train these neural networks, and it's fast, and it has an insatiable desire for data. So what we've worked with NetApp is actually pool together reference architecture so that when a data scientist, who is a very valuable resource, is working on this, he's ensured that the infrastructures are gonna work together seamlessly and deliver that data to the training process. And then when you create that model, we use something that's called inference, you put it in production, and again same time, when you're having that inference running you wanna make sure that data can get to it and can interact with the data seamlessly and the reference architectures play out there as well. So our goal is, start knocking off one by one, what do the customers need to be successful? And we put a lot of effort into the GPUs, we put a lot of effort into the deep learning software that runs on top of that, we put a lot of effort into, you know, what's the models they need to use, etc. And now we have to spend a lot more time of what's their infrastructure? And make sure that's reliable because, you would hate to do all that work only to find that your infrastructure had a hiccup, and took your job down. So we're working really hard to make sure that never happens >> So I have this theory that, well I don't have the theory, David Curry came out with this theory of data has gravity, but I've come up with this additional theory, now that we look at AI, and the capability of AI and what people are and what the hyper scalers are doing in their data center is that individual companies think, have a challenge replicating in their own data center, this AI and compute now has gravity. You know, I can't well, at least before today I didn't think well I can take my data center, put it on the road, and do these massive pieces of injection on the edge, sounds like we're pushin' back on that a little bit and saying that you know what sure if it's, I don't know what the limits are, and I guess that's the question. What are the limits of what we can do on the edge when it comes to the amount of data, and portable AI to that edge? >> Well so, there's again the two aspects of it, the training takes an incredible amount of data that's why they would have to take a super computer and put it there so they could do the retraining, but, when you think about when you can have the pro-- something the size of a credit card, which is our Jetson solution, and you can install it in a drone or you can put in cameras for public safety, etc. Which is, has incredible, think about looking for a lost child or parents with Alzheimer's, you can scan through video real quick and find them, right? All because of a credit card sized processor, that's pretty impressive. But that's what's happening at the edge, we're now writing applications that are much more intelligent using AI, there are AI applications sitting at the edge that, instead of just processing the data in a way where I'm getting a average, average number of people who walked into my store, right, that's what we used to do five years ago, now we're actually using intelligent applications that are making calculated decisions, it's understanding who's coming in a store, understanding their buying/purchasing power, etc. That's extremely important in retail, because, if you wanna interact with someone and give them that, you know when they're doing self checkout, try to sell 'em one more thing, you know, did you forget the batteries that go with that, or whatever you want it to be, you only have a few seconds, right? And so you must be able to process that and have something really intelligent doing that instead of just trying to do the law of average and get a directionally correct-- and we've known this, anytime you've been on your webpage or whatever and someone recommends something you're like that doesn't have anything to do with me and then all of a sudden it started getting really good that's where they're getting more intelligent. >> When I walk into the store with my White Sox hat and then they recommend the matching jersey. I'm gonna look, gonna come lookin' for you guys at NVIDIA like wa-hey! I don't have money for a jersey, but things like that, yeah. >> We're just behind the scenes somewhere. >> Well, you title VP and GM of Deep Learning and stuff, there's a lot of stuff. (all laugh) Jim thanks so much for coming back on theCUBE sharing with us what's new at NVIDIA it sounds like the world of possibilities is endless, so exciting! >> Yeah, it is an exciting time, thank you. >> Thanks for your time, we wanna thank you for watching theCUBE, Lisa Martin with Keith Townsend from SAP SAPPHIRE 2018, thanks for watching. (bubbly music)

(music playing) >> Narrator: Live, from downtown San Francisco It's the Cube. Covering IBM Chief Data Officer Startegy Summit 2018. Brought to you by: IBM >> Welcome back to San Francisco everybody we're at the Parc 55 in Union Square. My name is Dave Vellante, and you're watching the Cube. The leader in live tech coverage and this is our exclusive coverage of IBM's Chief Data Officer Strategy Summit. They hold these both in San Francisco and in Boston. It's an intimate event, about 150 Chief Data Officers really absorbing what IBM has done internally and IBM transferring knowledge to its clients. Steven Eluk is here. He is one of those internal practitioners at IBM. He's the Vice President of Deep Learning and the Global Chief Data Office at IBM. We just heard from him and some of his strategies and used cases. He's joined by Sumit Gupta, a Cube alum. Who is the Vice President of Machine Learning and deep learning within IBM's cognitive systems group. Sumit. >> Thank you. >> Good to see you, welcome back Steven, lets get into it. So, I was um paying close attention when Bob Picciano took over the cognitive systems group. I said, "Hmm, that's interesting". Recently a software guy, of course I know he's got some hardware expertise. But bringing in someone who's deep into software and machine learning, and deep learning, and AI, and cognitive systems into a systems organization. So you guys specifically set out to develop solutions to solve problems like Steven's trying to solve. Right, explain that. >> Yeah, so I think ugh there's a revolution going on in the market the computing market where we have all these new machine learning, and deep learning technologies that are having meaningful impact or promise of having meaningful impact. But these new technologies, are actually significantly I would say complex and they require very complex and high performance computing systems. You know I think Bob and I think in particular IBM saw the opportunity and realized that we really need to architect a new class of infrastructure. Both software and hardware to address what data scientist like Steve are trying to do in the space, right? The open source software that's out there: Denzoflo, Cafe, Torch - These things are truly game changing. But they also require GPU accelerators. They also require multiple systems like... In fact interestingly enough you know some of the super computers that we've been building for the scientific computing world, those same technologies are now coming into the AI world and the enterprise. >> So, the infrastructure for AI, if I can use that term? It's got to be flexible, Steven we were sort of talking about that elastic versus I'm even extending it to plastic. As Sumit you just said, it's got to have that tooling, got to have that modern tooling, you've got to accommodate alternative processor capabilities um, and so, that forms what you've used Steven to sort of create new capabilities new business capabilities within IBM. I wanted to, we didn't touch upon this before, but we touched upon your data strategy before but tie it back to the line of business. You essentially are a presume a liaison between the line of business and the chief data office >> Steven: Yeah. >> Officer office. How did that all work out, and shake out? Did you defining the business outcomes, the requirements, how did you go about that? >> Well, actually, surprisingly, we have very little new use cases that we're generating internally from my organization. Because there's so many to pick from already throughout the organization, right? There's all these business units coming to us and saying, "Hey, now the data is in the data lake and now we know there's more data, now we want to do this. How do we do it?" You know, so that's where we come in, that's where we start touching and massaging and enabling them. And that's the main efforts that we have. We do have some derivative works that have come out, that have been like new offerings that you'll see here. But mostly we already have so many use cases that from those businesses units that we're really trying to heighten and bring extra value to those domains first. >> So, a lot of organizations sounds like IBM was similar you created the data lake you know, things like "a doop" made a lower cost to just put stuff in the data lake. But then, it's like "okay, now what?" >> Steven: Yeah. >> So is that right? So you've got the data and this bog of data and you're trying to make more sense out of it but get more value out of it? >> Steven: Absolutely. >> That's what they were pushing you to do? >> Yeah, absolutely. And with that, with more data you need more computational power. And actually Sumit and I go pretty far back and I can tell you from my previous roles I heightened to him many years ago some of the deficiencies in the current architecture in X86 etc and I said, "If you hit these points, I will buy these products." And what they went back and they did is they, they addressed all of the issues that I had. Like there's certain issues... >> That's when you were, sorry to interrupt, that's when you were a customer, right? >> Steven: That's when I was... >> An external customer >> Outside. I'm still an internal customer, so I've always been a customer I guess in that role right? >> Yep, yep. >> But, I need to get data to the computational device as quickly as possible. And with certain older gen technologies, like PTI Gen3 and certain issues around um x86. I couldn't get that data there for like high fidelity imaging for autonomous vehicles for ya know, high fidelity image analysis. But, with certain technologies in power we have like envy link and directly to the CPU. And we also have PTI Gen4, right? So, so these are big enablers for me so that I can really keep the utilization of those very expensive compute devices higher. Because they're not starved for data. >> And you've also put a lot of emphasis on IO, right? I mean that's... >> Yeah, you know if I may break it down right there's actually I would say three different pieces to the puzzle here right? The highest level from Steve's perspective, from Steven's teams perspective or any data scientist perspective is they need to just do their data science and not worry about the infrastructure, right? They actually don't want to know that there's an infrastructure. They want to say, "launch job" - right? That's the level of grand clarity we want, right? In the background, they want our schedulers, our software, our hardware to just seamlessly use either one system or scale to 100 systems, right? To use one GPU or to use 1,000 GPUs, right? So that's where our offerings come in, right. We went and built this offering called Powder and Powder essentially is open source software like TensorFlow, like Efi, like Torch. But performace and capabilities add it to make it much easier to use. So for example, we have an extremely terrific scheduling software that manages jobs called Spectrum Conductor for Spark. So as the name suggests, it uses Apache Spark. But again the data scientist doesn't know that. They say, "launch job". And the software actually goes and scales that job across tens of servers or hundreds of servers. The IT team can determine how many servers their going to allocate for data scientist. They can have all kinds of user management, data management, model management software. We take the open source software, we package it. You know surprisingly ugh most people don't realize this, the open source software like TensorFlow has primarily been built on a (mumbles). And most of our enterprise clients, including Steven, are on Redhat. So we, we engineered Redhat to be able to manage TensorFlow. And you know I chose those words carefully, there was a little bit of engineering both on Redhat and on TensorFlow to make that whole thing work together. Sounds trivial, took several months and huge value proposition to the enterprise clients. And then the last piece I think that Steven was referencing too, is we also trying to go and make the eye more accessible for non data scientist or I would say even data engineers. So we for example, have a software called Powder Vision. This takes images and videos, and automatically creates a trained deep learning model for them, right. So we analyze the images, you of course have to tell us in these images, for these hundred images here are the most important things. For example, you've identified: here are people, here are cars, here are traffic signs. But if you give us some of that labeled data, we automatically do the work that a data scientist would have done, and create this pre trained AI model for you. This really enables many rapid prototyping for a lot of clients who either kind of fought to have data scientists or don't want to have data scientists. >> So just to summarize that, the three pieces: It's making it simpler for the data scientists, just run the job - Um, the backend piece which is the schedulers, the hardware, the software doing its thing - and then its making that data science capability more accessible. >> Right, right, right. >> Those are the three layers. >> So you know, I'll resay it in my words maybe >> Yeah please. >> Ease of use right, hardware software optimized for performance and capability, and point and click AI, right. AI for non data scientists, right. It's like the three levels that I think of when I'm engaging with data scientists and clients. >> And essentially it's embedded AI right? I've been making the point today that a lot of the AI is going to be purchased from companies like IBM, and I'm just going to apply it. I'm not going to try to go build my own, own AI right? I mean, is that... >> No absolutely. >> Is that the right way to think about it as a practitioner >> I think, I think we talked about it a little bit about it on the panel earlier but if we can, if we can leverage these pre built models and just apply a little bit of training data it makes it so much easier for the organizations and so much cheaper. They don't have to invest in a crazy amount of infrastructure, all the labeling of data, they don't have to do that. So, I think it's definitely steering that way. It's going to take a little bit of time, we have some of them there. But as we as we iterate, we are going to get more and more of these types of you know, commodity type models that people could utilize. >> I'll give you an example, so we have a software called Intelligent Analytics at IBM. It's very good at taking any surveillance data and for example recognizing anomalies or you know if people aren't suppose to be in a zone. Ugh and we had a client who wanted to do worker safety compliance. So they want to make sure workers are wearing their safety jackets and their helmets when they're in a construction site. So we use surveillance data created a new AI model using Powder AI vision. We were then able to plug into this IVA - Intelligence Analytic Software. So they have the nice gooey base software for the dashboards and the alerts, yet we were able to do incremental training on their specific use case, which by the way, with their specific you know equipment and jackets and stuff like that. And create a new AI model, very quickly. For them to be able to apply and make sure their workers are actually complaint to all of the safety requirements they have on the construction site. >> Hmm interesting. So when I, Sometimes it's like a new form of capture says identify "all the pictures with bridges", right that's the kind of thing you're capable to do with these video analytics. >> That's exactly right. You, every, clients will have all kinds of uses I was at a, talking to a client, who's a major car manufacturer in the world and he was saying it would be great if I could identify the make and model of what cars people are driving into my dealership. Because I bet I can draw a ugh corelation between what they drive into and what they going to drive out of, right. Marketing insights, right. And, ugh, so there's a lot of things that people want to do with which would really be spoke in their use cases. And build on top of existing AI models that we have already. >> And you mentioned, X86 before. And not to start a food fight but um >> Steven: And we use both internally too, right. >> So lets talk about that a little bit, I mean where do you use X86 where do you use IBM Cognitive and Power Systems? >> I have a mix of both, >> Why, how do you decide? >> There's certain of work loads. I will delegate that over to Power, just because ya know they're data starved and we are noticing a complication is being impacted by it. Um, but because we deal with so many different organizations certain organizations optimize for X86 and some of them optimize for power and I can't pick, I have to have everything. Just like I mentioned earlier, I also have to support cloud on prim, I can't pick just to be on prim right, it so. >> I imagine the big cloud providers are in the same boat which I know some are your customers. You're betting on data, you're betting on digital and it's a good bet. >> Steven: Yeah, 100 percent. >> We're betting on data and AI, right. So I think data, you got to do something with the data, right? And analytics and AI is what people are doing with that data we have an advantage both at the hardware level and at the software level in these two I would say workloads or segments - which is data and AI, right. And we fundamentally have invested in the processor architecture to improve the performance and capabilities, right. You could offer a much larger AI models on a power system that you use than you can on an X86 system that you use. Right, that's one advantage. You can train and AI model four times faster on a power system than you can on an Intel Based System. So the clients who have a lot of data, who care about how fast their training runs, are the ones who are committing to power systems today. >> Mmm.Hmm. >> Latency requirements, things like that, really really big deal. >> So what that means for you as a practitioner is you can do more with less or is it I mean >> I can definitely do more with less, but the real value is that I'm able to get an outcome quicker. Everyone says, "Okay, you can just roll our more GPU's more GPU's, but run more experiments run more experiments". No no that's not actually it. I want to reduce the time for a an experiment Get it done as quickly as possible so I get that insight. 'Cause then what I can do I can get possibly cancel out a bunch of those jobs that are already running cause I already have the insight, knowing that that model is not doing anything. Alright, so it's very important to get the time down. Jeff Dean said it a few years ago, he uses the same slide often. But, you know, when things are taking months you know that's what happened basically from the 80's up until you know 2010. >> Right >> We didn't have the computation we didn't have the data. Once we were able to get that experimentation time down, we're able to iterate very very quickly on this. >> And throwing GPU's at the problem doesn't solve it because it's too much complexity or? >> It it helps the problem, there's no question. But when my GPU utilization goes from 95% down to 60% ya know I'm getting only a two-thirds return on investment there. It's a really really big deal, yeah. >> Sumit: I mean the key here I think Steven, and I'll draw it out again is this time to insight. Because time to insight actually is time to dollars, right. People are using AI either to make more money, right by providing better customer products, better products to the customers, giving better recommendations. Or they're saving on their operational costs right, they're improving their efficiencies. Maybe their routing their trucks in the right way, their routing their inventory in the right place, they're reducing the amount of inventory that they need. So in all cases you can actually coordinate AI to a revenue outcome or a dollar outcome. So the faster you can do that, you know, I tell most people that I engage with the hardware and software they get from us pays for itself very quickly. Because they make that much more money or they save that much more money, using power systems. >> We, we even see this internally I've heard stories and all that, Sumit kind of commented on this but - There's actually sales people that take this software & hardware out and they're able to get an outcome sometimes in certain situations where they just take the clients data and they're sales people they're not data scientists they train it it's so simple to use then they present the client with the outcomes the next day and the client is just like blown away. This isn't just a one time occurrence, like sales people are actually using this right. So it's getting to the area that it's so simple to use you're able to get those outcomes that we're even seeing it you know deals close quicker. >> Yeah, that's powerful. And Sumit to your point, the business case is actually really easy to make. You can say, "Okay, this initiative that you're driving what's your forecast for how much revenue?" Now lets make an assumption for how much faster we're going to be able to deliver it. And if I can show them a one day turn around, on a corpus of data, okay lets say two months times whatever, my time to break. I can run the business case very easily and communicate to the CFO or whomever the line of business head so. >> That's right. I mean just, I was at a retailer, at a grocery store a local grocery store in the bay area recently and he was telling me how In California we've passed legislation that does not allow plastic bags anymore. You have to pay for it. So people are bringing their own bags. But that's actually increased theft for them. Because people bring their own bag, put stuff in it and walk out. And he didn't want to have an analytic system that can detect if someone puts something in a bag and then did not buy it at purchase. So it's, in many ways they want to use the existing camera systems they have but automatically be able to detect fraudulent behavior or you know anomalies. And it's actually quite easy to do with a lot of the software we have around Power AI Vision, around video analytics from IBM right. And that's what we were talking about right? Take existing trained AI models on vision and enhance them for your specific use case and the scenarios you're looking for. >> Excellent. Guys we got to go. Thanks Steven, thanks Sumit for coming back on and appreciate the insights. >> Thank you >> Glad to be here >> You're welcome. Alright, keep it right there buddy we'll be back with our next guest. You're watching "The Cube" at IBM's CDO Strategy Summit from San Francisco. We'll be right back. (music playing)

>> Announcer: Live from downtown San Francisco, it's theCUBE covering IBM Chief Data Officer Strategy Summit 2018, brought to you by IBM. (upbeat music) >> Welcome back to San Francisco, everybody. You're watching theCUBE, the leader in live tech coverage. We're covering the IBM Chief Data Officer Strategy Summit #ibmcdo. Ed Walsh is here. He's the General Manager of IBM Storage, and Steven Eliuk who's the Vice President of Deep Learning in the Global Chief Data Office at IBM, Steven. >> Yes, sir. >> Good to see you again. Welcome to The CUBE. >> Pleasure to be here. So there's a great story. We heard Inderpal Bhandari this morning talk about the enterprise data blueprint and laying out to the practitioners how to get started, how to implement, and we're going to have a little case study as to actually how you're doing this. But Ed, set it up for us. >> Okay, so we're at this Chief Data Officer Summit in the Spring, we do it twice a year and really get just Chief Data Officers together to think through their different challenges and actually share. So that's where we're at the Summit. And what we've, as IBM, as kind of try to be a foot forward, be that cognitive enterprise and showing very transparently what we're doing at our organization be more data-driven. And we've talked a bunch of different times. Everyone needs to be data-driven. Everyone wants to be data-driven, but it's really challenging for organizations. So what we're doing is with this blueprint which we're showing as a showcase, in fact you can actually physically come in and see our environment. But more importantly we're being very transparent on all the different components, high-level processes, what we did in governance, but also down to the Lilly Technology level and sharing that with our... Not because they want to do all of it, but maybe they want to do some of it or half of it, but it would be a blueprint that's worked. And then we're being transparent about what we're getting internally for our own transformation as IBM. Because really if we looked at this as a platform, it's really an enterprise cognitive data platform that all of IBM uses on all our transformation work. So our client, in fact, is Steven, and I think you can give what are we doing. By the way, it also, same type of infrastructure allows you to do what we did in the national labs, the largest supercomputers in the world, same infrastructure and the same thing we're trying to do, is make it easier for people to get insights from the data at scale in the enterprise. So that's why I want to bring Steven on. >> I joked with Inderpal. I said, "Well, if you can do it at IBM, "if you can do it there you can do it anywhere," (Ed laughing) because he's point oh. We're at a highly complex organization. So Steven, take us through how you got started and what you're doing. >> For sure, so I'm what's referred to probably as a difficult customer. So because we're so multifaceted we have so many different use cases internally in the orders of hundreds, it doesn't mean that I can just say, "Hey, this is a specific pattern that I need, Ed. "You need to make sure your hardware is sufficient in this area," because the next day I'm going to be hitting him and say, "Hey Ed, I need you to make sure "that it's also efficient in terms of bandwidth as well." And that's the beauty of working in this domain, is that I have those hundreds of use cases and it means that I'm hitting low latency requirements, bandwidth requirements, extensibility requirements because I have a huge number of headcount that I'm bringing on as well. And if I'm good now I don't have to worry about in six months to be stating, "Hey, I need to roll out new infrastructure "so I can support these new data scientists "and effectively so that they can get outcomes quicker." And I'd need to make sure that all the infrastructure behind the scenes is extensible and supports my users. And what I don't want them to have to worry about specifically is how that infrastructure works. I want them to focus on those use cases, those enterprise use cases, and I want them to touch as many of those use cases as possible. >> So Inderpal laid out sort of his five things that a CDO should do. He starts with develop a clear data strategy. So as the doer in the organization, how'd you go about doing that? Presumably you participated in that data strategy, but you're representing the lines of business presumably to make sure that it's of value to them. You can accelerate business value, but how did you start? I mean that's a big challenge, chewy. >> For sure, yeah, it's a huge challenge. And I think effectively curating, locating, governing, and quality aspects of that data is one of the first aspects. And where does that data reside, though, and how do we access it quickly? How does it support structured and unstructured data effectively? Those are all really important questions that had to come to light. And that's some of the approaches that we took. We look at the various business units and we look at are they curating the data correctly? Is it the data that we need? Maybe we have to augment that curation process before we actually are able to kind of apply new techniques, new machine-learning techniques, to that use case. There's a number of different aspects that kind of get rolled into that, and bringing effective storage and effective compute to the table really accelerates us in that journey. >> So Ed, what are the fundamental aspects of the infrastructure that supports this sort of emerging workload? >> Yeah, no, good question. And some of it is what we're going to talk about, what's a storage layer and what's a compute layer, but also what are the tools we're putting in place to use a lot of these open-source toolsets and make it easier for people to use but also use that underlying infrastructure better. So if you look at the high level, we use a storage infrastructure that is built for these AI workloads which is closer to an HPC workload. So the same infrastructure we use, we use the term ESS or elastic storage server. It's a combination. It's a turnkey solution, half rack, full rack. But it can start very small and grow to the biggest supercomputers in the world like what we're doing in the national labs, like the largest top five supercomputers in the world. But what that is is a file system called Spectrum Scale. Allows you to scale up at the performance but also low latency, gets added to the metadata but also high throughput. So we can do layers on that either on flash being all the hot tiers'll be on flash because it's not just the throughput you need which is high. So our lowest end box's close to like what, 26 gigabytes a second. Our highest one like national labs is 4.9 terabytes a second throughput. But it's also the low latency quick access. So we have a storage infrastructure but then we also have high-performance compute. So what we have is our Power Systems, our POWER9 Systems with GPUs, and the idea is how do you, we use the term feed the beast? How do you have the right throughput or IOPS to get the data close to that CPU or the GPU? The Power Systems have a unique bandwidth, so it's not like what you just find from a Comodo, the Intel servers. It's a much faster throughput, so it allows us to actually get data between the GPU CPU in storage or memory very fast. So you can get these deep learning times, and maybe you can share some of that. The learning times go up dramatically, so you get the insight. And then we're also putting layers on top which are IBM Cloud Private, is basically how do you have a hybrid cloud container-based service that allows you to move things seamlessly across and not have to wrestle with how to put all these things together either so it works seamlessly between a public cloud and private cloud? Then we have these toolsets, and I talked about this last time. It might not seem like storage or what you have in APU but we use the term PowerAI, is taking all these machine-learning tools because everyone always used open source. But we make them one more scale but also to ease your use. So how do you use a bunch of great GPUs and CPUs, great throughput, and how do you scale that? A lot of these tools were basically to be run on one CPU. So to be distributed, key research from IBM allows you to actually with PowerAI take the same TensorFlow workflows or dot dot dot and run it across a grid dramatically changing what you're doing from learning times. But anyway you can probably give more, I think, but it's a multiple layer. It's not one thing but it's not what you use for digital storage infrastructure, compute infrastructure for normal workloads. It is custom so you can't... A lot of people try to deploy maybe their NAS storage box and maybe it's flash and try to deploy it. And you can get going that way but then you hit a wall real quick. This is purposely built for AI. >> So Beth Smith was on earlier. She threw out a stat. She said that 85% of their, based on some research, I'm not sure if it was IBM or Forrest or Gartner, said 85% of customers they talked to said AI will be a competitive advantage but only 20% can use it today at scale. So obviously scale is a big challenge, and I want to ask you to comment on another potential challenge. We always talk about elastic infrastructure. You scale up, scale down, or end of month, okay. We sometimes use this concept of plastic infrastructure. Basically plastic maintains its shape because these workloads are so diverse. I don't want to have to rip down my infrastructure and bring in a new one every time my workload changes. So I wonder if you can talk about the sort of requirements from your perspective both in terms of scale and in terms of adaptability to changing workloads. >> Well, I think one of the things that Ed brought up that's really, really important is these open-source frameworks assume that it's running on a single system. They assume that storage is actually local, and that's really the only way that you get really effective throughput from it, is if it's local. So extending it via PowerAI, via these appliances and so forth means that you can use petabytes of storage at a distance and still have good throughput and not have those GP utilization coming down because these are very expensive devices. So if the storage is the blocker, is their controller and he's limiting that flow of data then ultimately you're not making the most effective use of those very expensive computational mediums. But more importantly it means that your time from ideation to product is slowed down, so you're not able to get those business outcomes. That means your competitor could get those business outcomes if they don't have it. And for me what's really important is I mentioned this briefly earlier, is that I need those specialists to touch as much of the data or as much as those enterprise use cases as possible. At the end of the year it's not about touching three use cases. It's the touching three this year, five, ten, more and more and more. And with the infrastructure being storage and computation, all of that is key attributes to kind of seeing that goal. >> Without having to rip that down and then repurpose building it every time. >> Steven: Yeah. >> And just being able to deal with the grid as a grid and you can place workloads across a grid. >> 100%. >> That's our Spectrum compute products that we've been doing for all the major banks in the world to do that and take these workloads and place them across a grid is also a key piece of this. So we always talk about the infrastructures being hey, Ed, that's not storage or infrastructure. No, you need that. And that's why it's part of my portfolio to actually build out the overall infrastructure for people to build on prim but also talk about everything we did with you on prim is hybrid. It's goes to the Cloud natively because some workloads we believe will be on the Cloud for good reasons, and you need to have that part of it. So everything we're going with you is hybrid cloud today, not in the future, today. >> No, 100%, and that's one of the requirements in our organization that we call A-1 architecture. If we write it for our own prim we have to be able to run it on the Cloud and it has to have the same look and feel and painted glass and things like that as well. So it means we only have to write it once, so we're incredibly efficient because we don't have to write it multiple times for different types of infrastructure. Likewise we have expectations from the data scientists that the performance all still have to be up to par as well. We want to really be moving the computation directly to where the data resides and we know that it's not just on prim, it's not in the Cloud, it's a hybrid scenario. >> So don't hate me for asking you this, Ed, but you've only been here for a couple years. Did you just stumble into this? You got this vast portfolio, you got this tooling, you got cloud. You got a part of your organization saying we got to do on prim. The other part's saying we got to do public. Or was this designed to the workload? Was kind of a little bit of both? >> Well, I think luck is good, but it's a embarrassment of riches inside IBM between our primary research, some of the things we were just talking about. How do you run these frameworks in a distributed fashion and not designed that way and do it performing at scale? That's our primary, that's research. That's not even in my group. What we're doing is for workload management. That's in storage, but we have these toolsets. The key thing is work with the clients to figure out what they're trying to do. Everyone's trying to be data-driven, so as we looked at what you need to do to be truly data-driven, it's not just having faster storage although that's important. It's not about the throughput or having to scale up. It's not about having just the CPUs. It's not just about having the open frameworks, but it's how to put that all together that we're invisible. In fact you said it earlier. He doesn't want his users to know at all what's underneath. He just wants to run their workload. You have people from my organization because I'm one of your customers. You're my customer but we go to you and say, "We're trying to use your platform "for a 360 view of the client," and our not data scientists, not data engineers, but ops team can use his platform. So anyway, so I actually think it's because IBM has its broad portfolio that we can bring together. And when IBM shows up which we're showing up in AI together in the Cloud, that's when you see something that we can truly do that you can't get from other organizations. And it's because of the technology differentiation we have from the different groups, but also the industry contacts that we bring. >> 100%. >> And also when you're dealing with data it is the trust. We can engage the clients at a high level and help them because we're not a single-product company. We might be more complex, but when we show up and bring the solution set we can really differentiate. And I think that's when IBM shows up. It's pretty powerful. >> And I think it's moved from "trust me" as well to "show me," and we're able to show it now because we're eating what we're producing. So we're showing. They called it a blueprint. We're using that effectively inside the organization. >> So now that you've sort of built this out internally you spend a lot of time with clients kind of showing them or...? >> Probably 15% of my time. >> So not that much. >> No, no, because I'm in charge of internal transformation operations. They're expecting outcomes from us. But at the same time there's clients that are in the exact same boat. The realization that this is really interesting. There's a lot of noise, a lot of interesting stuff in AI out there from Google, from Facebook, from Amazon, from all, Microsoft, but image recognition isn't important to me. How do I do it for my own organization? I have legacy data from 50 years. This is totally different, and there's no Git repo that I can go to and download them all and use it. It's totally custom, and how do I handle that? So it's different for these guys. >> What's on your wishlist? What's on Ed's to do list? >> Oh geez, uh... I want it so simple for my data scientists that they don't have to worry about where the data's coming from. Whether it be a traditional relational database or an object store, I want it to feed that data effectively and I don't want to have to have them looking into where the data is to make sure the computation's there. I want it just to flow effortlessly. That's really the wishlist. Likewise, I think if we had new accelerators in general outside the box, not something from the traditional GPU viewpoint, maybe data flow or something in new avant-garde-type stuff, that would be interesting because I think it might open up a new train of thought in the area just like GPUs did for us. >> Great story. >> Yeah I know, I think it's... So we're talking about AI for business, and I think what you're seeing is we're trying to showcase what IBM's doing to be really an AI business. And what we've done in this platform is really a showcase. So we're trying to be as transparent as possible not because it's the only way to do it but it's a good example of how a very complex business is using AI to get dramatically better and everyone's using the same kind of platform. >> Well, we learned, we effectively learned being open is much better than being closed. Look at the AI community. Because of its openness that's where we're at right now. And following the same lead we're doing the same thing, and that's why we're making everything available. You can see it and we're doing it, and we're happy to talk to you about it. >> Awesome, all right, so Steven, you stay here. >> Yeah. >> We're going to bring Sumit on and we're going to drill down into the cognitive platform. >> That's good. This guy, thanks for setting it up. I really, really appreciate it. >> Thank you very much. >> All right, good having you guys. All right, keep it right there, everybody. We'll be back at the IBM CDO Strategy Summit. You're watching theCUBE. (upbeat music) (telephone dialing) (modem connecting)

>> Narrator: Live from Las Vegas, it's theCUBE. Covering AWS re:Invent 2017 presented by AWS, Intel and our ecosystem of partners. >> Hello and welcome back. Live here is theCUBE's exclusive coverage here in Las Vegas. 45,000 people attending Amazon Web Services' AWS re:Invent 2017. I'm John Furrier with Lisa Martin. Our next guest is Miles Kingston, he is the General Manager of the Smart Home Group at Intel Corporation. Miles, it's great to have you. >> Thank you so much for having me here, I'm really happy to be here. >> Welcome to theCUBE Alumni Club. First time on. All the benefits you get as being an Alumni is to come back again. >> Can't wait, I'll be here next year, for sure. >> Certainly, you are running a new business for Intel, I'd like to get some details on that, because smart homes. We were at the Samsung Developer Conference, we saw smart fridge, smart living room. So we're starting to see this become a reality, for the CES, every 10 years, that's smart living room. So finally, with cloud and all of the computing power, it's arrived or has it? >> I believe we're almost there. I think the technology has finally advanced enough and there is so much data available now that you have this combination of this technology that can analyze all of this data and truly start doing some of the artificial intelligence that will help you make your home smarter. >> And we've certainly seen the growth of Siri with Apple, Alexa for the home with Amazon, just really go crazy. In fact, during the Industry Day, yesterday, you saw the repeat session most attended by developers, was Alexa. So Alexa's got the minds and has captured the imagination of the developers. Where does it go from here and what is the difference between a smart home and a connected home? Can you just take a minute to explain and set the table on that? >> Yeah and I agree, the voice capability in the home, it's absolutely foundational. I think I saw a recent statistic that by 2022, 55% of US households are expected to have a smart speaker type device in their home. So that's a massive percentage. So I think, if you look in the industry, connected home and smart home, they're often use synonymously. We personally look at it as an evolution. And so what I mean by that is, today, we think the home is extremely connected. If I talk about my house, and I'm a total geek about this stuff, I've got 60 devices connected to an access point, I've got another 60 devices connected to an IOT hub. My home does not feel very smart. It's crazy connected, I can turn on lights on and off, sprinklers on and off, it's not yet smart. What we're really focused on at Intel, is accelerating that transition for your home to truly become a smart home and not just a connected home. >> And software is a key part of it, and I've seen developers attack this area very nicely. At the same time, the surface area with these Smart Homes for security issues, hackers. Cause WiFi is, you can run a process on, these are computers. So how does security fit into all of this? >> Yeah, security is huge and so at Intel we're focused on four technology pillars, which we'll get through during this discussion. One of the first ones is connectivity, and we actually have technology that goes into a WiFi access point, the actual silicon. It's optimized for many clients to be in the home, and also, we've partnered with companies, like McAfee, on security software that will sit on top of that. That will actually manage all of the connected devices in your home, as that extra layer of security. So we fundamentally agree that the security is paramount. >> One of the things that I saw on the website that says, Intel is taking a radically different approach based on proactive research into ways to increase smart home adoption. What makes Intel's approach radically different? >> Yeah, so I'm glad that you asked that. We've spent years going into thousands of consumers' homes in North America, Western Europe, China, etc. To truly understand some of the pain points they were experiencing. From that, we basically, gave all this information to our architects and we really synthesized it into what areas we need to advance technology to enable some of these richer use cases. So we're really working on those foundational building blocks and so those four ones I mentioned earlier, connectivity, that one is paramount. You know, if you want to add 35 to 100 devices in your home, you better make sure they're all connected, all the time and that you've got good bandwidth between them. The second technology was voice, and it's not just voice in one place in your home, it's voice throughout your home. You don't want to have to run to the kitchen to turn your bedroom lights on. And then, vision. You know, making sure your home has the ability to see more. It could be cameras, could be motion sensors, it could be vision sensors. And then this last one is this local intelligence. This artificial intelligence. So the unique approach that Intel is taking is across all of our assets. In the data center, in our artificial intelligence organization, in our new technology organization, our IOT organization, in our client computing group. We're taking all of these assets and investing them in those four pillars and kind of really delivering unique solutions, and there's actually a couple of them that have been on display this week so far. >> How about DeepLens? That certainly was an awesome keynote point, and the device that Andy introduced is essentially a wireless device, that is basically that machine learning an AI in it. And that is awesome, because it's also an IOT device, it's got so much versatility to it. What's behind that? Can you give some color to DeepLens? What does it mean for people? >> So, we're really excited about that one. We partnered with Amazon at AWS on that for quite some time. So, just a reminder to everybody, that is the first Deep Learning enabled wireless camera. And what we're helped do in that, is it's got an Intel Atom processor inside that actually runs the vision processing workload. We also contributed a Deep Learning toolkit, kind of a software middleware layer, and we've also got the Intel Compute Library for deep neural networks. So basically, a lot of preconfigured algorithms that developers can use. The bigger thing, though, is when I talked about those four technology pillars; the vision pillar, as well as the artificial intelligence pillar, this is a proof point of exactly that. Running an instance of the AWS service on a local device in the home to do this computer vision. >> When will that device be available? And what's the price point? Can we get our hands on one? And how are people going to be getting this? >> Yeah, so what was announced during the keynote today is that there are actually some Deep Learning workshops today, here at re:Invent where they're actually being given away, and then actually as soon as the announcement was made during the keynote today, they're actually available for pre-order on Amazon.com right now. I'm not actually sure on the shipping date on Amazon, but anybody can go and check. >> Jeff Frick, go get one of those quickly. Order it, put my credit card down. >> Miles: Yes, please do. >> Well, that's super exciting and now, where's the impact in that? Because it seems like it could be a great IOT device. It seems like it would be a fun consumer device. Where do you guys see the use cases for these developing? >> So the reason I'm excited about this one, is I fundamentally believe that vision is going to enable some richer use cases. The only way we're going to get those though, is if you get these brilliant developers getting their hands on the hardware, with someone like Amazon, who's made all of the machine learning, and the cloud and all of the pieces easier. It's now going to make it very easy for thousands, ideally, hundreds of thousands of developers to start working on this, so they can enable these new use cases. >> The pace of innovation that AWS has set, it's palpable here, we hear it, we feel it. This is a relatively new business unit for Intel. You announced this, about a year ago at re:Invent 2016? Are you trying to match the accelerated pace of innovation that AWS has? And what do you see going on in the next 12 months? Where do you think we'll be 12 months from now? >> Yeah, so I think we're definitely trying to be a fantastic technology partner for Amazon. One of the things we have since last re:Invent is we announced we were going to do some reference designs and developer kits to help get Alexa everywhere. So during this trade show, actually, we are holding, I can't remember the exact number, but many workshops, where we are providing the participants with a Speech Enabling Developer toolkit. And basically, what this is, is it's got an Intel platform, with Intel's dual DSP on it, a microarray, and it's paired with Raspberry Pi. So basically, this will allow anybody who already makes a product, it will allow them to easily integrate Alexa into that product with Intel inside. Which is perfect for us. >> So obviously, we're super excited, we love the cloud. I'm kind of a fanboy of the cloud, being a developer in my old days, but the resources that you get out of the cloud are amazing. But now when you start looking at these devices like DeepLens, the possibilities are limitless. So it's really interesting. The question I have for you is, you know, we had Tom Siebel on earlier, pioneer, invented the CRM category. He's now the CEO of C3 IOT, and I asked him, why are you doing a startup, you're a billionaire. You're rich, you don't need to do it. He goes, "I'm a computer guy, I love doing this." He's an entrepreneur at heart. But he said something interesting, he said that the two waves that he surfs, they call him a big time surfer, he's hanging 10 on the waves, is IOT and AI. This is an opportunity for you guys to reimagine the smart home. How important is the IOT trend and the AI trend for really doing it right with smart home, and whatever we're calling it. There's an opportunity there. How are you guys viewing that vision? What progress points have you identified at Intel, to kind of, check? >> Completely agree. For me, AI really is the key turning point here. 'Cause even just talking about connected versus smart, the thing that makes it smart is the ability to learn and think for itself. And the reason we have focused on those technology pillars, is we believe that by adding voice everywhere in the home, and the listening capability, as well as adding the vision capability, you're going to enable all of this rich new data, which you have to have some of these AI tools to make any sense of, and when you get to video, you absolutely have to have some amount of it locally. So, that either for bandwidth reasons, for latency reasons, for privacy reasons, like some of the examples that were given in the keynote today, you just want to keep that stuff locally. >> And having policy and running on it, you know, access points are interesting, it gives you connectivity, but these are computers, so if someone gets malware on the home, they can run a full threaded process on these machines. Sometimes you might not want that. You want to be able to control that. >> Yes, absolutely. We would really believe that the wireless access point in the home is one of the greatest areas where you can add additional security in the home and protect all of the devices. >> So you mentioned, I think 120 different devices in your home that are connected. How far away do you think your home is from being, from going from connected to smart? What's that timeline like? >> You know what I think, honestly, I think a lot of the hardware is already there. And the examples I will give is, and I'm not just saying this because I'm here, but I actually do have 15 Echos in my house because I do want to be able to control all of the infrastructure everywhere in the home. I do believe in the future, those devices will be listening for anomalies, like glass breaking, a dog barking, a baby crying, and I believe the hardware we have today is very capable of doing that. Similarly, I think that a lot of the cameras today are trained to, whenever they see motion, to do certain things and to start recording. I think that use case is going to evolve over time as well, so I truly believe that we are probably two years away from really seeing, with some of the existing infrastructure, truly being able to enable some smarter home use cases. >> The renaissance going on, the creativity is going to be amazing. I'm looking at a tweet that Bert Latimar, from our team made, on our last interview with the Washington County Sheriff, customer of Amazon, pays $6 a month for getting all the mugshots. He goes, "I'm gonna use DeepLens for things like "recognizing scars and tattoos." Because now they have to take pictures when someone comes in as a criminal, but now with DeepLens, they can program it to look for tattoos. >> Yeah, absolutely. And if you see things like the Ring Doorbell today, they have that neighborhood application of it so you can actually share within your local neighborhood if somebody had a package stolen, they can post a picture of that person. And even just security cameras, my house, it feels like Fort Knox sometimes, I've got so many security cameras. It used to be, every time there was a windstorm, I got 25 alerts on my phone, because a branch was blowing. Now I have security cameras that actually can do facial recognition and say, your son is home, your daughter is home, your wife is home. >> So are all the houses going to have a little sign that says,"Protected by Alexa and Intel and DeepLens" >> Don't you dare, exactly. (laughs) >> Lisa: And no sneaking out for the kids. >> Yes, exactly. >> Alright, so real quick to end the segment, quickly summarize and share, what is the Intel relationship with Amazon Web Services? Talk about the partnership. >> It's a great relationship. We've been partnering with Amazon for over a decade, starting with AWS. Over the last couple of years, we've started working closely with them on their first party products. So, many of you have seen the Echo Show and the Echo Look, that has Intel inside. It also has a RealSense Camera in the Look. We've now enabled the Speech Enabling Developer Kit, which is meant to help get Alexa everywhere, running on Intel. We've now done DeepLens, which is a great example of local artificial intelligence. Partnered with all the work we've done with them in the cloud, so it really is, I would say the partnership expands all the way from the very edge device in the home, all the way to the cloud. >> Miles, thanks for coming, Miles Kingston with Intel, General Manager of the Smart Home Group, new business unit at Intel, really reimagining the future for people's lives. I think in this great case where technology can actually help people, rather than making it any more complicated. Which we all know if we have access points and kids gaming, it can be a problem. It's theCUBE, live here in Las Vegas. 45,000 people here at Amazon re:Invent. Five years ago, our first show, only 7,000. Now what amazing growth. Thanks so much for coming out, Lisa Martin and John Furrier here, reporting from theCUBE. More coverage after this short break. (light music)

>> Announcer: From Denver, Colorado, it's theCUBE, covering Super Computing 17. Brought to you by Intel. (soft electronic music) Hey, welcome back, everybody. Jeff Frick here with theCUBE. We're gettin' towards the end of the day here at Super Computing 2017 in Denver, Colorado. 12,000 people talkin' really about the outer limits of what you can do with compute power and lookin' out into the universe and black holes and all kinds of exciting stuff. We're kind of bringin' it back, right? We're all about democratization of technology for people to solve real problems. We're really excited to have our last guest of the day, bringin' the energy, Armughan Ahmad. He's SVP and GM, Hybrid Cloud and Ready Solutions for Dell EMC, and a many-time CUBE alumni. Armughan, great to see you. >> Yeah, good to see you, Jeff. So, first off, just impressions of the show. 12,000 people, we had no idea. We've never been to this show before. This is great. >> This is a show that has been around. If you know the history of the show, this was an IEEE engineering show, that actually turned into high-performance computing around research-based analytics and other things that came out of it. But, it's just grown. We're seeing now, yesterday the super computing top petaflops were released here. So, it's fascinating. You have some of the brightest minds in the world that actually come to this event. 12,000 of them. >> Yeah, and Dell EMC is here in force, so a lot of announcements, a lot of excitement. What are you guys excited about participating in this type of show? >> Yeah, Jeff, so when we come to an event like this, HBC-- We know that HBC is also evolved from your traditional HBC, which was around modeling and simulation, and how it started from engineering to then clusters. It's now evolving more towards machine learning, deep learning, and artificial intelligence. So, what we announced here-- Yesterday, our press release went out. It was really related to how our strategy of advancing HBC, but also democratizing HBC's working. So, on the advancing, on the HBC side, the top 500 super computing list came out. We're powering some of the top 500 of those. One big one is TAC, which is Texas Institute out of UT, University of Texas. They now have, I believe, the number 12 spot in the top 500 super computers in the world, running an 8.2 petaflops off computing. >> So, a lot of zeros. I have no idea what a petaflop is. >> It's very, very big. It's very big. It's available for machine learning, but also eventually going to be available for deep learning. But, more importantly, we're also moving towards democratizing HBC because we feel that democratizing is also very important, where HBC should not only be for the research and the academia, but it should also be focused towards the manufacturing customers, the financial customers, our commercial customers, so that they can actually take the complexity of HBC out, and that's where our-- We call it our HBC 2.0 strategy, off learning from the advancements that we continue to drive, to then also democratizing it for our customers. >> It's interesting, I think, back to the old days of Intel microprocessors getting better and better and better, and you had Spark and you had Silicon Graphics, and these things that were way better. This huge differentiation. But, the Intel I32 just kept pluggin' along and it really begs the question, where is the distinction now? You have huge clusters of computers you can put together with virtualization. Where is the difference between just a really big cluster and HBC and super computing? >> So, I think, if you look at HBC, HBC is also evolving, so let's look at the customer view, right? So, the other part of our announcement here was artificial intelligence, which is really, what is artificial intelligence? It's, if you look at a customer retailer, a retailer has-- They start with data, for example. You buy beer and chips at J's Retailer, for example. You come in and do that, you usually used to run a SEQUEL database or you used to run a RDBMS database, and then that would basically tell you, these are the people who can purchase from me. You know their purchase history. But, then you evolved into BI, and then if that data got really, very large, you then had an HBC cluster, would which basically analyze a lot of that data for you, and show you trends and things. That would then tell you, you know what, these are my customers, this is how many times they are frequent. But, now it's moving more towards machine learning and deep learning as well. So, as the data gets larger and larger, we're seeing datas becoming larger, not just by social media, but your traditional computational frameworks, your traditional applications and others. We're finding that data is also growing at the edge, so by 2020, about 20 billion devices are going to wake up at the edge and start generating data. So, now, Internet data is going to look very small over the next three, four years, as the edge data comes up. So, you actually need to now start thinking of machine learning and deep learning a lot more. So, you asked the question, how do you see that evolving? So, you see an RDBMS traditional SQL evolving to BI. BI then evolves into either an HBC or hadoop. Then, from HBC and hadoop, what do you do next? What you do next is you start to now feed predictive analytics into machine learning kind of solutions, and then once those predictive analytics are there, then you really, truly start thinking about the full deep learning frameworks. >> Right, well and clearly like the data in motion. I think it's funny, we used to make decisions on a sample of data in the past. Now, we have the opportunity to take all the data in real time and make those decisions with Kafka and Spark and Flink and all these crazy systems that are comin' to play. Makes Hadoop look ancient, tired, and yesterday, right? But, it's still valid, right? >> A lot of customers are still paying. Customers are using it, and that's where we feel we need to simplify the complex for our customers. That's why we announced our Machine Learning Ready Bundle and our Deep Learning Ready Bundle. We announced it with Intel and Nvidia together, because we feel like our customers either go to the GPU route, which is your accelerator's route. We announced-- You were talking to Ravi, from our server team, earlier, where he talked about the C4140, which has the quad GPU power, and it's perfect for deep learning. But, with Intel, we've also worked on the same, where we worked on the AI software with Intel. Why are we doing all of this? We're saying that if you thought that RDBMS was difficult, and if you thought that building a hadoop cluster or HBC was a little challenging and time consuming, as the customers move to machine learning and deep learning, you now have to think about the whole stack. So, let me explain the stack to you. You think of a compute storage and network stack, then you think of-- The whole eternity. Yeah, that's right, the whole eternity of our data center. Then you talk about our-- These frameworks, like Theano, Caffe, TensorFlow, right? These are new frameworks. They are machine learning and deep learning frameworks. They're open source and others. Then you go to libraries. Then you go to accelerators, which accelerators you choose, then you go to your operating systems. Now, you haven't even talked about your use case. Retail use case or genomic sequencing use case. All you're trying to do is now figure out TensorFlow works with this accelerator or does not work with this accelerator. Or, does Caffe and Theano work with this operating system or not? And, that is a complexity that is way more complex. So, that's where we felt that we really needed to launch these new solutions, and we prelaunched them here at Super Computing, because we feel the evolution of HBC towards AI is happening. We're going to start shipping these Ready Bundles for machine learning and deep learning in first half of 2018. >> So, that's what the Ready Solutions are? You're basically putting the solution together for the client, then they can start-- You work together to build the application to fix whatever it is they're trying to do. >> That's exactly it. But, not just fix it. It's an outcome. So, I'm going to go back to the retailer. So, if you are the CEO of the biggest retailer and you are saying, hey, I just don't want to know who buys from me, I want to now do predictive analytics, which is who buys chips and beer, but who can I sell more things to, right? So, you now start thinking about demographic data. You start thinking about payroll data and other datas that surround-- You start feeding that data into it, so your machine now starts to learn a lot more of those frameworks, and then can actually give you predictive analytics. But, imagine a day where you actually-- The machine or the deep learning AI actually tells you that it's not just who you want to sell chips and beer to, it's who's going to buy the 4k TV? You're makin' a lot of presumptions. Well, there you go, and the 4k-- But, I'm glad you're doin' the 4k TV. So, that's important, right? That is where our customers need to understand how predictive analytics are going to move towards cognitive analytics. So, this is complex but we're trying to make that complex simple with these Ready Solutions from machine learning and deep learning. >> So, I want to just get your take on-- You've kind of talked about these three things a couple times, how you delineate between AI, machine learning, and deep learning. >> So, as I said, there is an evolution. I don't think a customer can achieve artificial intelligence unless they go through the whole crawl walk around space. There's no shortcuts there, right? What do you do? So, if you think about, Mastercard is a great customer of ours. They do an incredible amount of transactions per day, (laughs) as you can think, right? In millions. They want to do facial recognitions at kiosks, or they're looking at different policies based on your buying behavior-- That, hey, Jeff doesn't buy $20,000 Rolexes every year. Maybe once every week, you know, (laughs) it just depends how your mood is. I was in the Emirates. Exactly, you were in Dubai (laughs). Then, you think about his credit card is being used where? And, based on your behaviors that's important. Now, think about, even for Mastercard, they have traditional RDBMS databases. They went to BI. They have high-performance computing clusters. Then, they developed the hadoop cluster. So, what we did with them, we said okay. All that is good. That data that has been generated for you through customers and through internal IT organizations, those things are all very important. But, at the same time, now you need to start going through this data and start analyzing this data for predictive analytics. So, they had 1.2 million policies, for example, that they had to crunch. Now, think about 1.2 million policies that they had to say-- In which they had to take decisions on. That they had to take decisions on. One of the policies could be, hey, does Jeff go to Dubai to buy a Rolex or not? Or, does Jeff do these other patterns, or is Armughan taking his card and having a field day with it? So, those are policies that they feed into machine learning frameworks, and then machine learning actually gives you patterns that they can now see what your behavior is. Then, based on that, eventually deep learning is when they move to next. Deep learning now not only you actually talk about your behavior patterns on the credit card, but your entire other life data starts to-- Starts to also come into that. Then, now, you're actually talking about something before, that's for catching a fraud, you can actually be a lot more predictive about it and cognitive about it. So, that's where we feel that our Ready Solutions around machine learning and deep learning are really geared towards, so taking HBC to then democratizing it, advancing it, and then now helping our customers move towards machine learning and deep learning, 'cause these buzzwords of AIs are out there. If you're a financial institution and you're trying to figure out, who is that customer who's going to buy the next mortgage from you? Or, who are you going to lend to next? You want the machine and others to tell you this, not to take over your life, but to actually help you make these decisions so that your bottom line can go up along with your top line. Revenue and margins are important to every customer. >> It's amazing on the credit card example, because people get so pissed if there's a false positive. With the amount of effort that they've put into keep you from making fraudulent transactions, and if your credit card ever gets denied, people go bananas, right? The behavior just is amazing. But, I want to ask you-- We're comin' to the end of 2017, which is hard to believe. Things are rolling at Dell EMC. Michael Dell, ever since he took that thing private, you could see the sparkle in his eye. We got him on a CUBE interview a few years back. A year from now, 2018. What are we going to talk about? What are your top priorities for 2018? >> So, number one, Michael continues to talk about that our vision is advancing human progress through technology, right? That's our vision. We want to get there. But, at the same time we know that we have to drive IT transformation, we have to drive workforce transformation, we have to drive digital transformation, and we have to drive security transformation. All those things are important because lots of customers-- I mean, Jeff, do you know like 75% of the S&P 500 companies will not exist by 2027 because they're either not going to be able to make that shift from Blockbuster to Netflix, or Uber taxi-- It's happened to our friends at GE over the last little while. >> You can think about any customer-- That's what Michael did. Michael actually disrupted Dell with Dell technologies and the acquisition of EMC and Pivotal and VMWare. In a year from now, our strategy is really about edge to core to the cloud. We think the world is going to be all three, because the rise of 20 billion devices at the edge is going to require new computational frameworks. But, at the same time, people are going to bring them into the core, and then cloud will still exist. But, a lot of times-- Let me ask you, if you were driving an autonomous vehicle, do you want that data-- I'm an Edge guy. I know where you're going with this. It's not going to go, right? You want it at the edge, because data gravity is important. That's where we're going, so it's going to be huge. We feel data gravity is going to be big. We think core is going to be big. We think cloud's going to be big. And we really want to play in all three of those areas. >> That's when the speed of light is just too damn slow, in the car example. You don't want to send it to the data center and back. You don't want to send it to the data center, you want those decisions to be made at the edge. Your manufacturing floor needs to make the decision at the edge as well. You don't want a lot of that data going back to the cloud. All right, Armughan, thanks for bringing the energy to wrap up our day, and it's great to see you as always. Always good to see you guys, thank you. >> All right, this is Armughan, I'm Jeff Frick. You're watching theCUBE from Super Computing Summit 2017. Thanks for watching. We'll see you next time. (soft electronic music)

>> Announcer: Please welcome Vice President of Worldwide Marketing, Amazon Web Services, Ariel Kelman. (applause) (techno music) >> Good afternoon, everyone. Thank you for coming. I hope you guys are having a great day here. It is my pleasure to introduce to come up on stage here, the CEO of Amazon Web Services, Andy Jassy. (applause) (techno music) >> Okay. Let's get started. I have a bunch of questions here for you, Andy. >> Just like one of our meetings, Ariel. >> Just like one of our meetings. So, I thought I'd start with a little bit of a state of the state on AWS. Can you give us your quick take? >> Yeah, well, first of all, thank you, everyone, for being here. We really appreciate it. We know how busy you guys are. So, hope you're having a good day. You know, the business is growing really quickly. In the last financials, we released, in Q four of '16, AWS is a 14 billion dollar revenue run rate business, growing 47% year over year. We have millions of active customers, and we consider an active customer as a non-Amazon entity that's used the platform in the last 30 days. And it's really a very broad, diverse customer set, in every imaginable size of customer and every imaginable vertical business segment. And I won't repeat all the customers that I know Werner went through earlier in the keynote, but here are just some of the more recent ones that you've seen, you know NELL is moving their their digital and their connected devices, meters, real estate to AWS. McDonalds is re-inventing their digital platform on top of AWS. FINRA is moving all in to AWS, yeah. You see at Reinvent, Workday announced AWS was its preferred cloud provider, and to start building on top of AWS further. Today, in press releases, you saw both Dunkin Donuts and Here, the geo-spatial map company announced they'd chosen AWS as their provider. You know and then I think if you look at our business, we have a really large non-US or global customer base and business that continues to expand very dramatically. And we're also aggressively increasing the number of geographic regions in which we have infrastructure. So last year in 2016, on top of the broad footprint we had, we added Korea, India, and Canada, and the UK. We've announced that we have regions coming, another one in China, in Ningxia, as well as in France, as well as in Sweden. So we're not close to being done expanding geographically. And then of course, we continue to iterate and innovate really quickly on behalf of all of you, of our customers. I mean, just last year alone, we launched what we considered over 1,000 significant services and features. So on average, our customers wake up every day and have three new capabilities they can choose to use or not use, but at their disposal. You've seen it already this year, if you look at Chime, which is our new unified communication service. It makes meetings much easier to conduct, be productive with. You saw Connect, which is our new global call center routing service. If you look even today, you look at Redshift Spectrum, which makes it easy to query all your data, not just locally on disk in your data warehouse but across all of S3, or DAX, which puts a cash in front of DynamoDB, we use the same interface, or all the new features in our machine learning services. We're not close to being done delivering and iterating on your behalf. And I think if you look at that collection of things, it's part of why, as Gartner looks out at the infrastructure space, they estimate the AWS is several times the size business of the next 14 providers combined. It's a pretty significant market segment leadership position. >> You talked a lot about adopts in there, a lot of customers moving to AWS, migrating large numbers of workloads, some going all in on AWS. And with that as kind of backdrop, do you still see a role for hybrid as being something that's important for customers? >> Yeah, it's funny. The quick answer is yes. I think the, you know, if you think about a few years ago, a lot of the rage was this debate about private cloud versus what people call public cloud. And we don't really see that debate very often anymore. I think relatively few companies have had success with private clouds, and most are pretty substantially moving in the direction of building on top of clouds like AWS. But, while you increasingly see more and more companies every month announcing that they're going all in to the cloud, we will see most enterprises operate in some form of hybrid mode for the next number of years. And I think in the early days of AWS and the cloud, I think people got confused about this, where they thought that they had to make this binary decision to either be all in on the public cloud and AWS or not at all. And of course that's not the case. It's not a binary decision. And what we know many of our enterprise customers want is they want to be able to run the data centers that they're not ready to retire yet as seamlessly as they can alongside of AWS. And it's why we've built a lot of the capabilities we've built the last several years. These are things like PPC, which is our virtual private cloud, which allows you to cordon off a portion of our network, deploy resources into it and connect to it through VPN or Direct Connect, which is a private connection between your data centers and our regions or our storage gateway, which is a virtual storage appliance, or Identity Federation, or a whole bunch of capabilities like that. But what we've seen, even though the vast majority of the big hybrid implementations today are built on top of AWS, as more and more of the mainstream enterprises are now at the point where they're really building substantial cloud adoption plans, they've come back to us and they've said, well, you know, actually you guys have made us make kind of a binary decision. And that's because the vast majority of the world is virtualized on top of VMWare. And because VMWare and AWS, prior to a few months ago, had really done nothing to try and make it easy to use the VMWare tools that people have been using for many years seamlessly with AWS, customers were having to make a binary choice. Either they stick with the VMWare tools they've used for a while but have a really tough time integrating with AWS, or they move to AWS and they have to leave behind the VMWare tools they've been using. And it really was the impetus for VMWare and AWS to have a number of deep conversations about it, which led to the announcement we made late last fall of VMWare and AWS, which is going to allow customers who have been using the VMWare tools to manage their infrastructure for a long time to seamlessly be able to run those on top of AWS. And they get to do so as they move workloads back and forth and they evolve their hybrid implementation without having to buy any new hardware, which is a big deal for companies. Very few companies are looking to find ways to buy more hardware these days. And customers have been very excited about this prospect. We've announced that it's going to be ready in the middle of this year. You see companies like Amadeus and Merck and Western Digital and the state of Louisiana, a number of others, we've a very large, private beta and preview happening right now. And people are pretty excited about that prospect. So we will allow customers to run in the mode that they want to run, and I think you'll see a huge transition over the next five to 10 years. >> So in addition to hybrid, another question we get a lot from enterprises around the concept of lock-in and how they should think about their relationship with the vendor and how they should think about whether to spread the workloads across multiple infrastructure providers. How do you think about that? >> Well, it's a question we get a lot. And Oracle has sure made people care about that issue. You know, I think people are very sensitive about being locked in, given the experience that they've had over the last 10 to 15 years. And I think the reality is when you look at the cloud, it really is nothing like being locked into something like Oracle. The APIs look pretty similar between the various providers. We build an open standard, it's like Linux and MySQL and Postgres. All the migration tools that we build allow you to migrate in or out of AWS. It's up to customers based on how they want to run their workload. So it is much easier to move away from something like the cloud than it is from some of the old software services that has created some of this phobia. But I think when you look at most CIOs, enterprise CIOs particularly, as they think about moving to the cloud, many of them started off thinking that they, you know, very well might split their workloads across multiple cloud providers. And I think when push comes to shove, very few decide to do so. Most predominately pick an infrastructure provider to run their workloads. And the reason that they don't split it across, you know, pretty evenly across clouds is a few reasons. Number one, if you do so, you have to standardize in the lowest common denominator. And these platforms are in radically different stages at this point. And if you look at something like AWS, it has a lot more functionality than anybody else by a large margin. And we're also iterating more quickly than you'll find from the other providers. And most folks don't want to tie the hands of their developers behind their backs in the name of having the ability of splitting it across multiple clouds, cause they actually are, in most of their spaces, competitive, and they have a lot of ideas that they want to actually build and invent on behalf of their customers. So, you know, they don't want to actually limit their functionality. It turns out the second reason is that they don't want to force their development teams to have to learn multiple platforms. And most development teams, if any of you have managed multiple stacks across different technologies, and many of us have had that experience, it's a pain in the butt. And trying to make a shift from what you've been doing for the last 30 years on premises to the cloud is hard enough. But then forcing teams to have to get good at running across two or three platforms is something most teams don't relish, and it's wasteful of people's time, it's wasteful of natural resources. That's the second thing. And then the third reason is that you effectively diminish your buying power because all of these cloud providers have volume discounts, and then you're splitting what you buy across multiple providers, which gives you a lower amount you buy from everybody at a worse price. So when most CIOs and enterprises look at this carefully, they don't actually end up splitting it relatively evenly. They predominately pick a cloud provider. Some will just pick one. Others will pick one and then do a little bit with a second, just so they know they can run with a second provider, in case that relationship with the one they choose to predominately run with goes sideways in some fashion. But when you really look at it, CIOs are not making that decision to split it up relatively evenly because it makes their development teams much less capable and much less agile. >> Okay, let's shift gears a little bit, talk about a subject that's on the minds of not just enterprises but startups and government organizations and pretty much every organization we talk to. And that's AI and machine learning. Reinvent, we introduced our Amazon AI services and just this morning Werner announced the general availability of Amazon Lex. So where are we overall on machine learning? >> Well it's a hugely exciting opportunity for customers, and I think, we believe it's exciting for us as well. And it's still in the relatively early stages, if you look at how people are using it, but it's something that we passionately believe is going to make a huge difference in the world and a huge difference with customers, and that we're investing a pretty gigantic amount of resource and capability for our customers. And I think the way that we think about, at a high level, the machine learning and deep learning spaces are, you know, there's kind of three macro layers of the stack. I think at that bottom layer, it's generally for the expert machine learning practitioners, of which there are relatively few in the world. It's a scarce resource relative to what I think will be the case in five, 10 years from now. And these are folks who are comfortable working with deep learning engines, know how to build models, know how to tune those models, know how to do inference, know how to get that data from the models into production apps. And for that group of people, if you look at the vast majority of machine learning and deep learning that's being done in the cloud today, it's being done on top of AWS, are P2 instances, which are optimized for deep learning and our deep learning AMIs, that package, effectively the deep learning engines and libraries inside those AMIs. And you see companies like Netflix, Nvidia, and Pinterest and Stanford and a whole bunch of others that are doing significant amounts of machine learning on top of those optimized instances for machine learning and the deep learning AMIs. And I think that you can expect, over time, that we'll continue to build additional capabilities and tools for those expert practitioners. I think we will support and do support every single one of the deep learning engines on top of AWS, and we have a significant amount of those workloads with all those engines running on top of AWS today. We also are making, I would say, a disproportionate investment of our own resources and the MXNet community just because if you look at running deep learning models once you get beyond a few GPUs, it's pretty difficult to have those scale as you get into the hundreds of GPUs. And most of the deep learning engines don't scale very well horizontally. And so what we've found through a lot of extensive testing, cause remember, Amazon has thousands of deep learning experts inside the company that have built very sophisticated deep learning capabilities, like the ones you see in Alexa, we have found that MXNet scales the best and almost linearly, as we continue to add nodes, as we continue to horizontally scale. So we have a lot of investment at that bottom layer of the stack. Now, if you think about most companies with developers, it's still largely inaccessible to them to do the type of machine learning and deep learning that they'd really like to do. And that's because the tools, I think, are still too primitive. And there's a number of services out there, we built one ourselves in Amazon Machine Learning that we have a lot of customers use, and yet I would argue that all of those services, including our own, are still more difficult than they should be for everyday developers to be able to build machine learning and access machine learning and deep learning. And if you look at the history of what AWS has done, in every part of our business, and a lot of what's driven us, is trying to democratize technologies that were really only available and accessible before to a select, small number of companies. And so we're doing a lot of work at what I would call that middle layer of the stack to get rid of a lot of the muck associated with having to do, you know, building the models, tuning the models, doing the inference, figuring how to get the data into production apps, a lot of those capabilities at that middle layer that we think are really essential to allow deep learning and machine learning to reach its full potential. And then at the top layer of the stack, we think of those as solutions. And those are things like, pass me an image and I'll tell you what that image is, or show me this face, does it match faces in this group of faces, or pass me a string of text and I'll give you an mpg file, or give me some words and what your intent is and then I'll be able to return answers that allow people to build conversational apps like the Lex technology. And we have a whole bunch of other services coming in that area, atop of Lex and Polly and Recognition, and you can imagine some of those that we've had to use in Amazon over the years that we'll continue to make available for you, our customers. So very significant level of investment at all three layers of that stack. We think it's relatively early days in the space but have a lot of passion and excitement for that. >> Okay, now for ML and AI, we're seeing customers wanting to load in tons of data, both to train the models and to actually process data once they've built their models. And then outside of ML and AI, we're seeing just as much demand to move in data for analytics and traditional workloads. So as people are looking to move more and more data to the cloud, how are we thinking about making it easier to get data in? >> It's a great question. And I think it's actually an often overlooked question because a lot of what gets attention with customers is all the really interesting services that allow you to do everything from compute and storage and database and messaging and analytics and machine learning and AI. But at the end of the day, if you have a significant amount of data already somewhere else, you have to get it into the cloud to be able to take advantage of all these capabilities that you don't have on premises. And so we have spent a disproportionate amount of focus over the last few years trying to build capabilities for our customers to make this easier. And we have a set of capabilities that really is not close to matched anywhere else, in part because we have so many customers who are asking for help in this area that it's, you know, that's really what drives what we build. So of course, you could use the good old-fashioned wire to send data over the internet. Increasingly, we find customers that are trying to move large amounts of data into S3, is using our S3 transfer acceleration service, which basically uses our points of presence, or POPs, all over the world to expedite delivery into S3. You know, a few years ago, we were talking to a number of companies that were looking to make big shifts to the cloud, and they said, well, I need to move lots of data that just isn't viable for me to move it over the wire, given the connection we can assign to it. It's why we built Snowball. And so we launched Snowball a couple years ago, which is really, it's a 50 terabyte appliance that is encrypted, the data's encrypted three different ways, and you ingest the data from your data center into Snowball, it has a Kindle connected to it, it allows you to, you know, that makes sure that you send it to the right place, and you can also track the progress of your high-speed ingestion into our data centers. And when we first launched Snowball, we launched it at Reinvent a couple years ago, I could not believe that we were going to order as many Snowballs to start with as the team wanted to order. And in fact, I reproached the team and I said, this is way too much, why don't we first see if people actually use any of these Snowballs. And so the team thankfully didn't listen very carefully to that, and they really only pared back a little bit. And then it turned out that we, almost from the get-go, had ordered 10X too few. And so this has been something that people have used in a very broad, pervasive way all over the world. And last year, at the beginning of the year, as we were asking people what else they would like us to build in Snowball, customers told us a few things that were pretty interesting to us. First, one that wasn't that surprising was they said, well, it would be great if they were bigger, you know, if instead of 50 terabytes it was more data I could store on each device. Then they said, you know, one of the problems is when I load the data onto a Snowball and send it to you, I have to still keep my local copy on premises until it's ingested, cause I can't risk losing that data. So they said it would be great if you could find a way to provide clustering, so that I don't have to keep that copy on premises. That was pretty interesting. And then they said, you know, there's some of that data that I'd actually like to be loading synchronously to S3, and then, or some things back from S3 to that data that I may want to compare against. That was interesting, having that endpoint. And then they said, well, we'd really love it if there was some compute on those Snowballs so I can do analytics on some relatively short-term signals that I want to take action on right away. Those were really the pieces of feedback that informed Snowball Edge, which is the next version of Snowball that we launched, announced at Reinvent this past November. So it has, it's a hundred-terabyte appliance, still the same level of encryption, and it has clustering so that you don't have to keep that copy of the data local. It allows you to have an endpoint to S3 to synchronously load data back and forth, and then it has a compute inside of it. And so it allows customers to use these on premises. I'll give you a good example. GE is using these for their wind turbines. And they collect all kinds of data from those turbines, but there's certain short-term signals they want to do analytics on in as close to real time as they can, and take action on those. And so they use that compute to do the analytics and then when they fill up that Snowball Edge, they detach it and send it back to AWS to do broad-scale analytics in the cloud and then just start using an additional Snowball Edge to capture that short-term data and be able to do those analytics. So Snowball Edge is, you know, we just launched it a couple months ago, again, amazed at the type of response, how many customers are starting to deploy those all over the place. I think if you have exabytes of data that you need to move, it's not so easy. An exabyte of data, if you wanted to move from on premises to AWS, would require 10,000 Snowball Edges. Those customers don't want to really manage a fleet of 10,000 Snowball Edges if they don't have to. And so, we tried to figure out how to solve that problem, and it's why we launched Snowmobile back at Reinvent in November, which effectively, it's a hundred-petabyte container on a 45-foot trailer that we will take a truck and bring out to your facility. It comes with its own power and its own network fiber that we plug in to your data center. And if you want to move an exabyte of data over a 10 gigabit per second connection, it would take you 26 years. But using 10 Snowmobiles, it would take you six months. So really different level of scale. And you'd be surprised how many companies have exabytes of data at this point that they want to move to the cloud to get all those analytics and machine learning capabilities running on top of them. Then for streaming data, as we have more and more companies that are doing real-time analytics of streaming data, we have Kinesis, where we built something called the Kinesis Firehose that makes it really simple to stream all your real-time data. We have a storage gateway for companies that want to keep certain data hot, locally, and then asynchronously be loading the rest of their data to AWS to be able to use in different formats, should they need it as backup or should they choose to make a transition. So it's a very broad set of storage capabilities. And then of course, if you've moved a lot of data into the cloud or into anything, you realize that one of the hardest parts that people often leave to the end is ETL. And so we have announced an ETL service called Glue, which we announced at Reinvent, which is going to make it much easier to move your data, be able to find your data and map your data to different locations and do ETL, which of course is hugely important as you're moving large amounts. >> So we've talked a lot about moving things to the cloud, moving applications, moving data. But let's shift gears a little bit and talk about something not on the cloud, connected devices. >> Yeah. >> Where do they fit in and how do you think about edge? >> Well, you know, I've been working on AWS since the start of AWS, and we've been in the market for a little over 11 years at this point. And we have encountered, as I'm sure all of you have, many buzzwords. And of all the buzzwords that everybody has talked about, I think I can make a pretty strong argument that the one that has delivered fastest on its promise has been IOT and connected devices. Just amazing to me how much is happening at the edge today and how fast that's changing with device manufacturers. And I think that if you look out 10 years from now, when you talk about hybrid, I think most companies, majority on premise piece of hybrid will not be servers, it will be connected devices. There are going to be billions of devices all over the place, in your home, in your office, in factories, in oil fields, in agricultural fields, on ships, in cars, in planes, everywhere. You're going to have these assets that sit at the edge that companies are going to want to be able to collect data on, do analytics on, and then take action. And if you think about it, most of these devices, by their very nature, have relatively little CPU and have relatively little disk, which makes the cloud disproportionately important for them to supplement them. It's why you see most of the big, successful IOT applications today are using AWS to supplement them. Illumina has hooked up their genome sequencing to AWS to do analytics, or you can look at Major League Baseball Statcast is an IOT application built on top of AWS, or John Deer has over 200,000 telematically enabled tractors that are collecting real-time planting conditions and information that they're doing analytics on and sending it back to farmers so they can figure out where and how to optimally plant. Tata Motors manages their truck fleet this way. Phillips has their smart lighting project. I mean, there're innumerable amounts of these IOT applications built on top of AWS where the cloud is supplementing the device's capability. But when you think about these becoming more mission-critical applications for companies, there are going to be certain functions and certain conditions by which they're not going to want to connect back to the cloud. They're not going to want to take the time for that round trip. They're not going to have connectivity in some cases to be able to make a round trip to the cloud. And what they really want is customers really want the same capabilities they have on AWS, with AWS IOT, but on the devices themselves. And if you've ever tried to develop on these embedded devices, it's not for mere mortals. It's pretty delicate and it's pretty scary and there's a lot of archaic protocols associated with it, pretty tough to do it all and to do it without taking down your application. And so what we did was we built something called Greengrass, and we announced it at Reinvent. And Greengrass is really like a software module that you can effectively have inside your device. And it allows developers to write lambda functions, it's got lambda inside of it, and it allows customers to write lambda functions, some of which they want to run in the cloud, some of which they want to run on the device itself through Greengrass. So they have a common programming model to build those functions, to take the signals they see and take the actions they want to take against that, which is really going to help, I think, across all these IOT devices to be able to be much more flexible and allow the devices and the analytics and the actions you take to be much smarter, more intelligent. It's also why we built Snowball Edge. Snowball Edge, if you think about it, is really a purpose-built Greengrass device. We have Greengrass, it's inside of the Snowball Edge, and you know, the GE wind turbine example is a good example of that. And so it's to us, I think it's the future of what the on-premises piece of hybrid's going to be. I think there're going to be billions of devices all over the place and people are going to want to interact with them with a common programming model like they use in AWS and the cloud, and we're continuing to invest very significantly to make that easier and easier for companies. >> We've talked about several feature directions. We talked about AI, machine learning, the edge. What are some of the other areas of investment that this group should care about? >> Well there's a lot. (laughs) That's not a suit question, Ariel. But there's a lot. I think, I'll name a few. I think first of all, as I alluded to earlier, we are not close to being done expanding geographically. I think virtually every tier-one country will have an AWS region over time. I think many of the emerging countries will as well. I think the database space is an area that is radically changing. It's happening at a faster pace than I think people sometimes realize. And I think it's good news for all of you. I think the database space over the last few decades has been a lonely place for customers. I think that they have felt particularly locked into companies that are expensive and proprietary and have high degrees of lock-in and aren't so customer-friendly. And I think customers are sick of it. And we have a relational database service that we launched many years ago and has many flavors that you can run. You can run MySQL, you can run Postgres, you can run MariaDB, you can run SQLServer, you can run Oracle. And what a lot of our customers kept saying to us was, could you please figure out a way to have a database capability that has the performance characteristics of the commercial-grade databases but the customer-friendly and pricing model of the more open engines like the MySQL and Postgres and MariaDB. What you do on your own, we do a lot of it at Amazon, but it's hard, I mean, it takes a lot of work and a lot of tuning. And our customers really wanted us to solve that problem for them. And it's why we spent several years building Aurora, which is our own database engine that we built, but that's fully compatible with MySQL and with Postgres. It's at least as fault tolerant and durable and performant as the commercial-grade databases, but it's a tenth of the cost of those. And it's also nice because if it turns out that you use Aurora and you decide for whatever reason you don't want to use Aurora anymore, because it's fully compatible with MySQL and Postgres, you just dump it to the community versions of those, and off you are. So there's really hardly any transition there. So that is the fastest-growing service in the history of AWS. I'm amazed at how quickly it's grown. I think you may have heard earlier, we've had 23,000 database migrations just in the last year or so. There's a lot of pent-up demand to have database freedom. And we're here to help you have it. You know, I think on the analytic side, it's just never been easier and less expensive to collect, store, analyze, and share data than it is today. Part of that has to do with the economics of the cloud. But a lot of it has to do with the really broad analytics capability that we provide you. And it's a much broader capability than you'll find elsewhere. And you know, you can manage Hadoop and Spark and Presto and Hive and Pig and Yarn on top of AWS, or we have a managed elastic search service, and you know, of course we have a very high scale, very high performing data warehouse in Redshift, that just got even more performant with Spectrum, which now can query across all of your S3 data, and of course you have Athena, where you can query S3 directly. We have a service that allows you to do real-time analytics of streaming data in Kinesis. We have a business intelligence service in QuickSight. We have a number of machine learning capabilities I talked about earlier. It's a very broad array. And what we find is that it's a new day in analytics for companies. A lot of the data that companies felt like they had to throw away before, either because it was too expensive to hold or they didn't really have the tools accessible to them to get the learning from that data, it's a totally different day today. And so we have a pretty big investment in that space, I mentioned Glue earlier to do ETL on all that data. We have a lot more coming in that space. I think compute, super interesting, you know, I think you will find, I think we will find that companies will use full instances for many, many years and we have, you know, more than double the number of instances than you'll find elsewhere in every imaginable shape and size. But I would also say that the trend we see is that more and more companies are using smaller units of compute, and it's why you see containers becoming so popular. We have a really big business in ECS. And we will continue to build out the capability there. We have companies really running virtually every type of container and orchestration and management service on top of AWS at this point. And then of course, a couple years ago, we pioneered the event-driven serverless capability in compute that we call Lambda, which I'm just again, blown away by how many customers are using that for everything, in every way. So I think the basic unit of compute is continuing to get smaller. I think that's really good for customers. I think the ability to be serverless is a very exciting proposition that we're continuing to to fulfill that vision that we laid out a couple years ago. And then, probably, the last thing I'd point out right now is, I think it's really interesting to see how the basic procurement of software is changing. In significant part driven by what we've been doing with our Marketplace. If you think about it, in the old world, if you were a company that was buying software, you'd have to go find bunch of the companies that you should consider, you'd have to have a lot of conversations, you'd have to talk to a lot of salespeople. Those companies, by the way, have to have a big sales team, an expensive marketing budget to go find those companies and then go sell those companies and then both companies engage in this long tap-dance around doing an agreement and the legal terms and the legal teams and it's just, the process is very arduous. Then after you buy it, you have to figure out how you're going to actually package it, how you're deploy to infrastructure and get it done, and it's just, I think in general, both consumers of software and sellers of software really don't like the process that's existed over the last few decades. And then you look at AWS Marketplace, and we have 35 hundred product listings in there from 12 hundred technology providers. If you look at the number of hours, that software that's been running EC2 just in the last month alone, it's several hundred million hours, EC2 hours, of that software being run on top of our Marketplace. And it's just completely changing how software is bought and procured. I think that if you talk to a lot of the big sellers of software, like Splunk or Trend Micro, there's a whole number of them, they'll tell you it totally changes their ability to be able to sell. You know, one of the things that really helped AWS in the early days and still continues to help us, is that we have a self-service model where we don't actually have to have a lot of people talk to every customer to get started. I think if you're a seller of software, that's very appealing, to allow people to find your software and be able to buy it. And if you're a consumer, to be able to buy it quickly, again, without the hassle of all those conversations and the overhead associated with that, very appealing. And I think it's why the marketplace has just exploded and taken off like it has. It's also really good, by the way, for systems integrators, who are often packaging things on top of that software to their clients. This makes it much easier to build kind of smaller catalogs of software products for their customers. I think when you layer on top of that the capabilities that we've announced to make it easier for SASS providers to meter and to do billing and to do identity is just, it's a very different world. And so I think that also is very exciting, both for companies and customers as well as software providers. >> We certainly touched on a lot here. And we have a lot going on, and you know, while we have customers asking us a lot about how they can use all these new services and new features, we also tend to get a lot of questions from customers on how we innovate so quickly, and they can think about applying some of those lessons learned to their own businesses. >> So you're asking how we're able to innovate quickly? >> Mmm hmm. >> I think there's a few things that have helped us, and it's different for every company. But some of these might be helpful. I'll point to a few. I think the first thing is, I think we disproportionately index on hiring builders. And we think of builders as people who are inventors, people who look at different customer experiences really critically, are honest about what's flawed about them, and then seek to reinvent them. And then people who understand that launch is the starting line and not the finish line. There's very little that any of us ever built that's a home run right out of the gate. And so most things that succeed take a lot of listening to customers and a lot of experimentation and a lot of iterating before you get to an equation that really works. So the first thing is who we hire. I think the second thing is how we organize. And we have, at Amazon, long tried to organize into as small and separable and autonomous teams as we can, that have all the resources in those teams to own their own destiny. And so for instance, the technologists and the product managers are part of the same team. And a lot of that is because we don't want the finger pointing that goes back and forth between the teams, and if they're on the same team, they focus all their energy on owning it together and understanding what customers need from them, spending a disproportionate amount of time with customers, and then they get to own their own roadmaps. One of the reasons we don't publish a 12 to 18 month roadmap is we want those teams to have the freedom, in talking to customers and listening to what you tell us matters, to re-prioritize if there are certain things that we assumed mattered more than it turns out it does. So, you know I think that the way that we organize is the second piece. I think a third piece is all of our teams get to use the same AWS building blocks that all of you get to use, which allow you to move much more quickly. And I think one of the least told stories about Amazon over the last five years, in part because people have gotten interested in AWS, is people have missed how fast our consumer business at Amazon has iterated. Look at the amount of invention in Amazon's consumer business. And they'll tell you that a big piece of that is their ability to use the AWS building blocks like they do. I think a fourth thing is many big companies, as they get larger, what starts to happen is what people call the institutional no, which is that leaders walk into meetings on new ideas looking to find ways to say no, and not because they're ill intended but just because they get more conservative or they have a lot on their plate or things are really managed very centrally, so it's hard to imagine adding more to what you're already doing. At Amazon, it's really the opposite, and in part because of the way we're organized in such a decoupled, decentralized fashion, and in part because it's just part of our DNA. When the leaders walk into a meeting, they are looking for ways to say yes. And we don't say yes to everything, we have a lot of proposals. But we say yes to a lot more than I think virtually any other company on the planet. And when we're having conversations with builders who are proposing new ideas, we're in a mode where we're trying to problem-solve with them to get to yes, which I think is really different. And then I think the last thing is that we have mechanisms inside the company that allow us to make fast decisions. And if you want a little bit more detail, you should read our founder and CEO Jeff Bezos's shareholder letter, which just was released. He talks about the fast decision-making that happens inside the company. It's really true. We make fast decisions and we're willing to fail. And you know, we sometimes talk about how we're working on several of our next biggest failures, and we hope that most of the things we're doing aren't going to fail, but we know, if you're going to push the envelope and if you're going to experiment at the rate that we're trying to experiment, to find more pillars that allow us to do more for customers and allow us to be more relevant, you are going to fail sometimes. And you have to accept that, and you have to have a way of evaluating people that recognizes the inputs, meaning the things that they actually delivered as opposed to the outputs, cause on new ventures, you don't know what the outputs are going to be, you don't know consumers or customers are going to respond to the new thing you're trying to build. So you have to be able to reward employees on the inputs, you have to have a way for them to continue to progress and grow in their career even if they work on something didn't work. And you have to have a way of thinking about, when things don't work, how do I take the technology that I built as part of that, that really actually does work, but I didn't get it right in the form factor, and use it for other things. And I think that when you think about a culture like Amazon, that disproportionately hires builders, organizes into these separable, autonomous teams, and allows them to use building blocks to move fast, and has a leadership team that's looking to say yes to ideas and is willing to fail, you end up finding not only do you do more inventing but you get the people at every level of the organization spending their free cycles thinking about new ideas because it actually pays to think of new ideas cause you get a shot to try it. And so that has really helped us and I think most of our customers who have made significant shifts to AWS and the cloud would argue that that's one of the big transformational things they've seen in their companies as well. >> Okay. I want to go a little bit deeper on the subject of culture. What are some of the things that are most unique about the AWS culture that companies should know about when they're looking to partner with us? >> Well, I think if you're making a decision on a predominant infrastructure provider, it's really important that you decide that the culture of the company you're going to partner with is a fit for yours. And you know, it's a super important decision that you don't want to have to redo multiple times cause it's wasted effort. And I think that, look, I've been at Amazon for almost 20 years at this point, so I have obviously drank the Kool Aid. But there are a few things that I think are truly unique about Amazon's culture. I'll talk about three of them. The first is I think that we are unusually customer-oriented. And I think a lot of companies talk about being customer-oriented, but few actually are. I think most of the big technology companies truthfully are competitor-focused. They kind of look at what competitors are doing and then they try to one-up one another. You have one or two of them that I would say are product-focused, where they say, hey, it's great, you Mr. and Mrs. Customer have ideas on a product, but leave that to the experts, and you know, you'll like the products we're going to build. And those strategies can be good ones and successful ones, they're just not ours. We are driven by what customers tell us matters to them. We don't build technology for technology's sake, we don't become, you know, smitten by any one technology. We're trying to solve real problems for our customers. 90% of what we build is driven by what you tell us matters. And the other 10% is listening to you, and even if you can't articulate exactly what you want, trying to read between the lines and invent on your behalf. So that's the first thing. Second thing is that we are pioneers. We really like to invent, as I was talking about earlier. And I think most big technology companies at this point have either lost their will or their DNA to invent. Most of them acquire it or fast follow. And again, that can be a successful strategy. It's just not ours. I think in this day and age, where we're going through as big a shift as we are in the cloud, which is the biggest technology shift in our lifetime, as dynamic as it is, being able to partner with a company that has the most functionality, it's iterating the fastest, has the most customers, has the largest ecosystem of partners, has SIs and ISPs, that has had a vision for how all these pieces fit together from the start, instead of trying to patch them together in a following act, you have a big advantage. I think that the third thing is that we're unusually long-term oriented. And I think that you won't ever see us show up at your door the last day of a quarter, the last day of a year, trying to harass you into doing some kind of deal with us, not to be heard from again for a couple years when we either audit you or try to re-up you for a deal. That's just not the way that we will ever operate. We are trying to build a business, a set of relationships, that will outlast all of us here. And I think something that always ties it together well is this trusted advisor capability that we have inside our support function, which is, you know, we look at dozens of programmatic ways that our customers are using the platform and reach out to you if you're doing something we think's suboptimal. And one of the things we do is if you're not fully utilizing resources, or hardly, or not using them at all, we'll reach out and say, hey, you should stop paying for this. And over the last couple of years, we've sent out a couple million of these notifications that have led to actual annualized savings for customers of 350 million dollars. So I ask you, how many of your technology partners reach out to you and say stop spending money with us? To the tune of 350 million dollars lost revenue per year. Not too many. And I think when we first started doing it, people though it was gimmicky, but if you understand what I just talked about with regard to our culture, it makes perfect sense. We don't want to make money from customers unless you're getting value. We want to reinvent an experience that we think has been broken for the prior few decades. And then we're trying to build a relationship with you that outlasts all of us, and we think the best way to do that is to provide value and do right by customers over a long period of time. >> Okay, keeping going on the culture subject, what about some of the quirky things about Amazon's culture that people might find interesting or useful? >> Well there are a lot of quirky parts to our culture. And I think any, you know lots of companies who have strong culture will argue they have quirky pieces but I think there's a few I might point to. You know, I think the first would be the first several years I was with the company, I guess the first six years or so I was at the company, like most companies, all the information that was presented was via PowerPoint. And we would find that it was a very inefficient way to consume information. You know, you were often shaded by the charisma of the presenter, sometimes you would overweight what the presenters said based on whether they were a good presenter. And vice versa. You would very rarely have a deep conversation, cause you have no room on PowerPoint slides to have any depth. You would interrupt the presenter constantly with questions that they hadn't really thought through cause they didn't think they were going to have to present that level of depth. You constantly have the, you know, you'd ask the question, oh, I'm going to get to that in five slides, you want to do that now or you want to do that in five slides, you know, it was just maddening. And we would often find that most of the meetings required multiple meetings. And so we made a decision as a company to effectively ban PowerPoints as a communication vehicle inside the company. Really the only time I do PowerPoints is at Reinvent. And maybe that shows. And what we found is that it's a much more substantive and effective and time-efficient way to have conversations because there is no way to fake depth in a six-page narrative. So what we went to from PowerPoint was six-page narrative. You can write, have as much as you want in the appendix, but you have to assume nobody will read the appendices. Everything you have to communicate has to be done in six pages. You can't fake depth in a six-page narrative. And so what we do is we all get to the room, we spend 20 minutes or so reading the document so it's fresh in everybody's head. And then where we start the conversation is a radically different spot than when you're hearing a presentation one kind of shallow slide at a time. We all start the conversation with a fair bit of depth on the topic, and we can really hone in on the three or four issues that typically matter in each of these conversations. So we get to the heart of the matter and we can have one meeting on the topic instead of three or four. So that has been really, I mean it's unusual and it takes some time getting used to but it is a much more effective way to pay attention to the detail and have a substantive conversation. You know, I think a second thing, if you look at our working backwards process, we don't write a lot of code for any of our services until we write and refine and decide we have crisp press release and frequently asked question, or FAQ, for that product. And in the press release, what we're trying to do is make sure that we're building a product that has benefits that will really matter. How many times have we all gotten to the end of products and by the time we get there, we kind of think about what we're launching and think, this is not that interesting. Like, people are not going to find this that compelling. And it's because you just haven't thought through and argued and debated and made sure that you drew the line in the right spot on a set of benefits that will really matter to customers. So that's why we use the press release. The FAQ is to really have the arguments up front about how you're building the product. So what technology are you using? What's the architecture? What's the customer experience? What's the UI look like? What's the pricing dimensions? Are you going to charge for it or not? All of those decisions, what are people going to be most excited about, what are people going to be most disappointed by. All those conversations, if you have them up front, even if it takes you a few times to go through it, you can just let the teams build, and you don't have to check in with them except on the dates. And so we find that if we take the time up front we not only get the products right more often but the teams also deliver much more quickly and with much less churn. And then the third thing I'd say that's kind of quirky is it is an unusually truth-seeking culture at Amazon. I think we have a leadership principle that we say have backbone, disagree, and commit. And what it means is that we really expect people to speak up if they believe that we're headed down a path that's wrong for customers, no matter who is advancing it, what level in the company, everybody is empowered and expected to speak up. And then once we have the debate, then we all have to pull the same way, even if it's a different way than you were advocating. And I think, you always hear the old adage of where, two people look at a ceiling and one person says it's 14 feet and the other person says, it's 10 feet, and they say, okay let's compromise, it's 12 feet. And of course, it's not 12 feet, there is an answer. And not all things that we all consider has that black and white answer, but most things have an answer that really is more right if you actually assess it and debate it. And so we have an environment that really empowers people to challenge one another and I think it's part of why we end up getting to better answers, cause we have that level of openness and rigor. >> Okay, well Andy, we have time for one more question. >> Okay. >> So other than some of the things you've talked about, like customer focus, innovation, and long-term orientation, what is the single most important lesson that you've learned that is really relevant to this audience and this time we're living in? >> There's a lot. But I'll pick one. I would say I'll tell a short story that I think captures it. In the early days at Amazon, our sole business was what we called an owned inventory retail business, which meant we bought the inventory from distributors or publishers or manufacturers, stored it in our own fulfillment centers and shipped it to customers. And around the year 1999 or 2000, this third party seller model started becoming very popular. You know, these were companies like Half.com and eBay and folks like that. And we had a really animated debate inside the company about whether we should allow third party sellers to sell on the Amazon site. And the concerns internally were, first of all, we just had this fundamental belief that other sellers weren't going to care as much about the customer experience as we did cause it was such a central part of everything we did DNA-wise. And then also we had this entire business and all this machinery that was built around owned inventory business, with all these relationships with publishers and distributors and manufacturers, who we didn't think would necessarily like third party sellers selling right alongside us having bought their products. And so we really debated this, and we ultimately decided that we were going to allow third party sellers to sell in our marketplace. And we made that decision in part because it was better for customers, it allowed them to have lower prices, so more price variety and better selection. But also in significant part because we realized you can't fight gravity. If something is going to happen, whether you want it to happen or not, it is going to happen. And you are much better off cannibalizing yourself or being ahead of whatever direction the world is headed than you are at howling at the wind or wishing it away or trying to put up blockers and find a way to delay moving to the model that is really most successful and has the most amount of benefits for the customers in question. And that turned out to be a really important lesson for Amazon as a company and for me, personally, as well. You know, in the early days of doing Marketplace, we had all kinds of folks, even after we made the decision, that despite the have backbone, disagree and commit weren't really sure that they believed that it was going to be a successful decision. And it took several months, but thankfully we really were vigilant about it, and today in roughly half of the units we sell in our retail business are third party seller units. Been really good for our customers. And really good for our business as well. And I think the same thing is really applicable to the space we're talking about today, to the cloud, as you think about this gigantic shift that's going on right now, moving to the cloud, which is, you know, I think in the early days of the cloud, the first, I'll call it six, seven, eight years, I think collectively we consumed so much energy with all these arguments about are people going to move to the cloud, what are they going to move to the cloud, will they move mission-critical applications to the cloud, will the enterprise adopt it, will public sector adopt it, what about private cloud, you know, we just consumed a huge amount of energy and it was, you can see both in the results in what's happening in businesses like ours, it was a form of fighting gravity. And today we don't really have if conversations anymore with our customers. They're all when and how and what order conversations. And I would say that this going to be a much better world for all of us, because we will be able to build in a much more cost effective fashion, we will be able to build much more quickly, we'll be able to take our scarce resource of engineers and not spend their resource on the undifferentiated heavy lifting of infrastructure and instead on what truly differentiates your business. And you'll have a global presence, so that you have lower latency and a better end user customer experience being deployed with your applications and infrastructure all over the world. And you'll be able to meet the data sovereignty requirements of various locales. So I think it's a great world that we're entering right now, I think we're at a time where there's a lot less confusion about where the world is headed, and I think it's an unprecedented opportunity for you to reinvent your businesses, reinvent your applications, and build capabilities for your customers and for your business that weren't easily possible before. And I hope you take advantage of it, and we'll be right here every step of the way to help you. Thank you very much. I appreciate it. (applause) >> Thank you, Andy. And thank you, everyone. I appreciate your time today. >> Thank you. (applause) (upbeat music)