>> Live from San Jose, in the heart of Silicon Valley, it's theCUBE, overing DataWorks Summit 2018. Brought to you by Hortonworks. >> Welcome back everyone to day two of theCUBE's live coverage of DataWorks, here in San Jose, California. I'm your host, Rebecca Knight, along with my co-host James Kobielus. We have two guests on this panel today, we have Tim Vincent, he is the VP of Cognitive Systems Software at IBM, and Steve Roberts, who is the Offering Manager for Big Data on IBM Power Systems. Thanks so much for coming on theCUBE. >> Oh thank you very much. >> Thanks for having us. >> So we're now in this new era, this Cognitive Systems era. Can you set the scene for our viewers, and tell our viewers a little bit about what you do and why it's so important >> Okay, I'll give a bit of a background first, because James knows me from my previous role as, and you know I spent a lot of time in the data and analytics space. I was the CTO for Bob running the analytics group up 'til about a year and a half ago, and we spent a lot of time looking at what we needed to do from a data perspective and AI's perspective. And Bob, when he moved over to the Cognitive Systems, Bob Picciano who's my current boss, Bob asked me to move over and really start helping build, help to build out more of a software, and more of an AI focus, and a workload focus on how we thinking of the Power brand. So we spent a lot of time on that. So when you talk about cognitive systems or AI, what we're really trying to do is think about how you actually couple a combination of software, so co-optimize software space and the hardware space specific of what's needed for AI systems. Because the act of processing, the data processing, the algorithmic processing for AI is very, very different then what you would have for traditional data workload. So we're spending a lot of time thinking about how you actually co-optimize those systems so you can actually build a system that's really optimized for the demands of AI. >> And is this driven by customers, is this driven by just a trend that IBM is seeing? I mean how are you, >> It's a combination of both. >> So a lot of this is, you know, there's a lot of thought put into this before I joined the team. So there was a lot of good thinking from the Power brand, but it was really foresight on things like Moore's Law coming to an end of it's lifecycle right, and the ramifications to that. And at the same time as you start getting into things like narrow NATS and the floating point operations that you need to drive a narrow NAT, it was clear that we were hitting the boundaries. And then there's new technologies such as what Nvidia produces with with their GPUs, that are clearly advantageous. So there's a lot of trends that were comin' together the technical team saw, and at the same time we were seeing customers struggling with specific things. You know how to actually build a model if the training time is going to be weeks, and months, or let alone hours. And one of the scenarios I like to think about, I was probably showing my age a bit, but went to a school called University of Waterloo, and when I went to school, and in my early years, they had a batch based system for compilation and a systems run. You sit in the lab at night and you submit a compile job and the compile job will say, okay it's going to take three hours to compile the application, and you think of the productivity hit that has to you. And now you start thinking about, okay you've got this new skill in data scientists, which is really, really hard to find, they're very, very valuable. And you're giving them systems that take hours and weeks to do what the need to do. And you know, so they're trying to drive these models and get a high degree of accuracy in their predictions, and they just can't do it. So there's foresight on the technology side and there's clear demand on the customer side as well. >> Before the cameras were rolling you were talking about how the term data scientists and app developers is used interchangeably, and that's just wrong. >> And actually let's hear, 'cause I'd be in this whole position that I agree with it. I think it's the right framework. Data science is a team sport but application development has an even larger team sport in which data scientists, data engineers play a role. So, yeah we want to hear your ideas on the broader application development ecosystem, and where data scientists, and data engineers, and sort, fall into that broader spectrum. And then how IBM is supporting that entire new paradigm of application development, with your solution portfolio including, you know Power, AI on Power? >> So I think you used the word collaboration and team sport, and data science is a collaborative team sport. But you're 100% correct, there's also a, and I think it's missing to a great degree today, and it's probably limiting the actual value AI in the industry, and that's had to be data scientists and the application developers interact with each other. Because if you think about it, one of the models I like to think about is a consumer-producer model. Who consumes things and who produces things? And basically the data scientists are producing a specific thing, which is you know simply an AI model, >> Machine models, deep-learning models. >> Machine learning and deep learning, and the application developers are consuming those things and then producing something else, which is the application logic which is driving your business processes, and this view. So they got to work together. But there's a lot of confusion about who does what. You know you see people who talk with data scientists, build application logic, and you know the number of people who are data scientists can do that is, you know it exists, but it's not where the value, the value they bring to the equation. And the application developers developing AI models, you know they exist, but it's not the most prevalent form fact. >> But you know it's kind of unbalanced Tim, in the industry discussion of these role definitions. Quite often the traditional, you know definition, our sculpting of data scientist is that they know statistical modeling, plus data management, plus coding right? But you never hear the opposite, that coders somehow need to understand how to build statistical models and so forth. Do you think that the coders of the future will at least on some level need to be conversant with the practices of building,and tuning, or training the machine learning models or no? >> I think it's absolutely happen. And I will actually take it a step further, because again the data scientist skill is hard for a lot of people to find. >> Yeah. >> And as such is a very valuable skill. And what we're seeing, and we are actually one of the offerings that we're pulling out is something called PowerAI Vision, and it takes it up another level above the application developer, which is how do you actually really unlock the capabilities of AI to the business persona, the subject matter expert. So in the case of vision, how do you actually allow somebody to build a model without really knowing what a deep learning algorithm is, what kind of narrow NATS you use, how to do data preparation. So we build a tool set which is, you know effectively a SME tool set, which allows you to automatically label, it actually allows you to tag and label images, and then as you're tagging and labeling images it learns from that and actually it helps automate the labeling of the image. >> Is this distinct from data science experience on the one hand, which is geared towards the data scientists and I think Watson Analytics among your tools, is geared towards the SME, this a third tool, or an overlap. >> Yeah this is a third tool, which is really again one of the co-optimized capabilities that I talked about, is it's a tool that we built out that really is leveraging the combination of what we do in Power, the interconnect which we have with the GPU's, which is the NVLink interconnect, which gives us basically a 10X improvement in bandwidth between the CPU and GPU. That allows you to actually train your models much more quickly, so we're seeing about a 4X improvement over competitive technologies that are also using GPU's. And if we're looking at machine learning algorithms, we've recently come out with some technology we call Snap ML, which allows you to push machine learning, >> Snap ML, >> Yeah, it allows you to push machine learning algorithms down into the GPU's, and this is, we're seeing about a 40 to 50X improvement over traditional processing. So it's coupling all these capabilities, but really allowing a business persona to something specific, which is allow them to build out AI models to do recognition on either images or videos. >> Is there a pre-existing library of models in the solution that they can tap into? >> Basically it allows, it has a, >> Are they pre-trained? >> No they're not pre-trained models that's one of the differences in it. It actually has a set of models that allow, it picks for you, and actually so, >> Oh yes, okay. >> So this is why it helps the business persona because it's helping them with labeling the data. It's also helping select the best model. It's doing things under the covers to optimize things like hyper-parameter tuning, but you know the end-user doesn't have to know about all these things right? So you're tryin' to lift, and it comes back to your point on application developers, it allows you to lift the barrier for people to do these tasks. >> Even for professional data scientists, there may be a vast library of models that they don't necessarily know what is the best fit for the particular task. Ideally you should have, the infrastructure should recommend and choose, under various circumstances, the models, and the algorithms, the libraries, whatever for you for to the task, great. >> One extra feature of PowerAI Enterprises is that it does include a way to do a quick visual inspection of a models accuracy with a small data sample before you invest in scaling over a cluster or large data set. So you can get a visual indicator as to the, whether the models moving towards accuracy or you need to go and test an alternate model. >> So it's like a dashboard, of like Gini coefficients and all that stuff, okay. >> Exactly it gives you a snapshot view. And the other thing I was going to mention, you guys talked about application development, data scientists and of course a big message here at the conference is, you know data science meets big data and the work that Hortonworks is doing involving the notion of container support in YARN, GPU awareness in YARN, bringing data science experience, which you can include the PowerAI capability that Tim was talking about, as a workload tightly coupled with Hadoop. And this is where our Power servers are really built, not for just a monolithic building block that always has the same ratio of compute and storage, but fit for purpose servers that can address either GPU optimized workloads, providing the bandwidth enhancements that Tim talked about with the GPU, but also day-to-day servers, that can now support two terrabytes of memory, double the overall memory bandwidth on the box, 44 cores that can support up to 176 threads for parallelization of Spark workloads, Sequel workloads, distributed data science workloads. So it's really about choosing the combination of servers that can really mix this evolving workload need, 'cause a dupe isn't now just map produced, it's a multitude of workloads that you need to be able to mix and match, and bring various capabilities to the table for a compute, and that's where Power8, now Power9 has really been built for this kind of combination workloads where you can add acceleration where it makes sense, add big data, smaller core, smaller memory, where it makes sense, pick and choose. >> So Steve at this show, at DataWorks 2018 here in San Jose, the prime announcement, partnership announced between IBM and Hortonworks was IHAH, which I believe is IBM Host Analytics on Hortonworks. What I want to know is that solution that runs inside, I mean it runs on top of HDP 3.0 and so forth, is there any tie-in from an offering management standpoint between that and PowerAI so you can build models in the PowerAI environment, and then deploy them out to, in conjunction with the IHAH, is there, going forward, I mean just wanted to get a sense of whether those kinds of integrations. >> Well the same data science capability, data science experience, whether you choose to run it in the public cloud, or run it in private cloud monitor on prem, it's the same data science package. You know PowerAI has a set of optimized deep-learning libraries that can provide advantage on power, apply when you choose to run those deployments on our Power system alright, so we can provide additional value in terms of these optimized libraries, this memory bandwidth improvements. So really it depends upon the customer requirements and whether a Power foundation would make sense in some of those deployment models. I mean for us here with Power9 we've recently announced a whole series of Linux Power9 servers. That's our latest family, including as I mentioned, storage dense servers. The one we're showcasing on the floor here today, along with GPU rich servers. We're releasing fresh reference architecture. It's really to support combinations of clustered models that can as I mentioned, fit for purpose for the workload, to bring data science and big data together in the right combination. And working towards cloud models as well that can support mixing Power in ICP with big data solutions as well. >> And before we wrap, we just wanted to wrap. I think in the reference architecture you describe, I'm excited about the fact that you've commercialized distributed deep-learning for the growing number of instances where you're going to build containerized AI and distributing pieces of it across in this multi-cloud, you need the underlying middleware fabric to allow all those pieces to play together into some larger applications. So I've been following DDL because you've, research lab has been posting information about that, you know for quite a while. So I'm excited that you guys have finally commercialized it. I think there's a really good job of commercializing what comes out of the lab, like with Watson. >> Great well a good note to end on. Thanks so much for joining us. >> Oh thank you. Thank you for the, >> Thank you. >> We will have more from theCUBE's live coverage of DataWorks coming up just after this. (bright electronic music)

>> Announcer: Live from Munich, Germany; it's The Cube! Covering IBM Fast Track Your Data, brought to you by IBM. >> Welcome back to Munich, Germany everybody. This is The Cube, the leader in live tech coverage. We're covering Fast Track Your Data, IBM's signature moment here in Munich. Big themes around GDPR, data science, data science being a team sport. I'm Dave Vellante, I'm here with my co-host Jim Kobielus. Marc Altshuller is here, he's the general manager of IBM Business Analytics. Good to see you again Marc. >> Hey, always great to see you. Welcome, it's our first time together. >> Okay so we heard your key note, you were talking about the caveats of correlations, you were talking about rear view mirror analysis versus sort of looking forward, something that I've been sort of harping on for years. You know, I mean I remember the early days of "decision support" and the promises of 360 degree views of the customer, and predictive analytics, and I've always said it, "DSS really never lived up to that", y'know? "Will big data live up to that?" and we're kind of living that now, but what's your take on where we're at in this whole databean? >> I mean look, different customers are at different ends of the spectrum, but people are really getting value. They're becoming these data driven businesses. I like what Rob Thomas talked about on stage, right. Visiting companies a few years ago where they'd say "I'm not a technology company.". Now, how can you possibly say you're not a technology company, regardless of the industry. Your competitors will beat you if they are using data and you're not. >> Yeah, and everybody talks about digital transformation. And you hear that a lot at conferences, you guys haven't been pounding that theme, other than, y'know below the surface. And to us, digital means data, right? And if you're going to transform digitally, then it's all about the data, you mentioned data driven. What are you seeing, I mean most organizations in our view aren't "data driven" they're sort of reactive. Their CEO's maybe want to be data driven, maybe they're aboard conversations as to how to get there, but they're mostly focused on "Alright, how do we keep "the lights on, how do we meet our revenue targets, "how do we grow a little bit, and then whatever money "we have leftover we'll try to, y'know transform." What are you seeing? Is that changing? >> I would say, look I can give you an example right from my own space, the software space. For years we would have product managers, offering managers, maybe interviewing clients, on gut feel deciding what features to put at what priority within the next release. Now we have all these products instrumented behind the scenes with data, so we can literally see the friction points, the exit points, how frequently they come back, how long they're sessions are, we can even see them effectively graduating within the system where they continue to learn, and where they had shorter sessions, they're now going the longer sessions. That's really, really powerful for us in terms of trying to maximize our outcome from a software perspective. So that's where we kind of like, drink our own champagne. >> I got to ask you, so in around 2003, 2004 HBR had an article, front page y'know cover article of how "gut feel beats data and analytics", now this is 2003, 2004, software development as you know it's a lot of art involved, so my question is how are you doing? Is the data informing you in ways that are nonintuitive? And is it driving y'know, business outcomes for IBM? >> It is, look you see, I'll see like GM's of sports teams talking about maybe pushing back a little bit on the data. It's not all data driven, there's a little bit of gut, like is the guy going to, is he a checker in hockey or whatever that happens to be, and I would say, when you actually look at what's going on within baseball, and you look at the data, when you watch baseball growing up, the commentator might say something along the lines of "the pitcher has their stuff" right? "Does the pitcher have their stuff or not?". Now they literally know, the release point based on elevation, IOT within the state of the release point, the spin velocity of the ball, where they mathematically know "does the pitcher have their stuff?", are they hitting their locations? So all that stuff has all become data driven, and if you don't want to embrace it, you get beat, right? I mean even in baseball, I remember talking to one of these Moneyball type guys where I said like "Doesn't weather impact baseball?" And they're like "Yeah, we've looked at that, it absolutely impacts it." 'Cause you always hear of football and remember the old Peyton Manning thing? Don't play Peyton Manning in cold weather, don't bet on Peyton Manning in cold weather. So "I'm like isn't the same in baseball?", And he's like, absolutely it's the same in baseball, players preform different based on the climate. Do any mangers change their lineup based on that? Never. >> Speaking of HBR, I mean in the last few years there was also an article or two by Michael Shrage about the whole notion of real world experimentation and e-commerce, driven by data, y'know in line, to an operational process, like tuning the design iteratively of say, a shopping cart within your e-commerce environment, based on the stats on what work and what does not work. So, in many ways I mean AB testing, real world experimentation thrives on data science. Do you see AB testing becoming a standard business practice everywhere, or only in particular industries like you know, like the Wal-marts of the world? >> Yeah, look so, AB testing, multi-variant testing, they're pervasive, pretty much anyone who has a website ought to be doing this if they're not doing it already. Maybe some startups aren't quite into it. They prioritized in different spots, but mainstream fortune 500 companies are doing this, the tools have made it really easy. I would say, maybe the Achilles heel or the next frontier is, that is effectively saying, kind of creating one pattern of user, putting everyone in a single bucket, right? "Does this button perform better "when it's orange or when it's green? "Oh, it performs better orange." Really, does it perform well for every segmentation orange better than green or is it just a certain segmentation? So that next kind of frontier is going to be, how do we segment it, know a little bit more about you when you're coming in so that AB testing starts to build these kind of sub-profiles, sub-segmentation. >> Micro-segmentation, and of course, the end extreme of that dynamic is one-to-one personalization of experiences and engagements based on knowing 360 degrees about you and what makes you tick as well, so yeah. >> Altshuller: And add onto that context, right? You have your business, let's even keep it really simple, right, you've got your business life, you've got your social life, and your profile of what you're looking for when you're shopping your social life or something is very different than when you're shopping your business life. We have to personalize it to the idea where, I don't want to say schizophrenic but you do have multiple personalities from an online perspective, right? From a digital perspective it all depends in the moment, what is it that you're actually doing, right? And what are you, who are you acting for? >> Marc, I want to ask you, you're homies, your peeps are the business people. >> Yes. >> That's where you spend your time. I'm interested in the relationship between those business people and the data science teams. They're all, we all hear about how data science and unicorns are hard to find, difficult to get the skills, citizen data science is sort of a nirvana. But, how are you seeing businesses bring the domain expertise of the business and blending that with data science? >> So, they do it, I have some cautionary tales that I've experienced in terms of how they're doing it. They feel like, let's just assign the subject matter expert, they'll work with the data scientist, they'll give them context as they're doing their project, but unfortunately what I've seen time and time again, is that subject matter expert right out of the gate brings a tremendous amount of bias based on the types of analysis they've done in the past. >> Vellante: That's not how we do it here. >> Yeah, exactly, like "did you test this?". "Oh yeah, there's no correlation there, we've tried it." Well, just because there's no correlation, as I talked about onstage, doesn't mean it's not part of the pattern in terms of, like you don't want someone in there right off the bat dismissing things. So I always coach, when the business user subject matter experts become involved early, they have to be tremendously open-minded and not all of them can be. I like bringing them in later, because that data scientist, they are unbiased, like they see this data set, it doesn't mean anything to them, they're just numerically telling you what the data set says. Now the business user can then add some context, maybe they grabbed a field that really is an irrelevant field and they can give them that context afterwards. But we just don't want them shutting down, kind of roots, too early in the process. >> You know, we've been talking for a couple of years now within our community about this digital matrix, this digital fabric that's emerged and you're seeing these horizontal layers of technology, whether it's cloud or, you know, security, you all OAuth in with LinkedIn, Facebook, and Twitter. There's a data fabric that's emerging and you're seeing all these new business models, whether it's Uber or Airbnb or WAZE, et cetera, and then you see this blockbuster announcement last week, Amazon buying Whole Foods. And it's just fascinating to us and it's all about the data that a company like an Amazon can be a content company, could be a retail company, now it's becoming a grocer, you see Apple getting into financial services. So, you're seeing industries being able to traverse or companies being able traverse industries and it's all because of the data, so these conversations absolutely are going on in boardrooms. It's all about the digital transformation, the digital disruption, so how do you see, you know, your clients trying to take advantage of that or defend against that? >> Yeah look, I mean, you have to be proactive. You have to be willing to disrupt yourself in all these tech industries, it's just moving too quickly. I read a similar story, I think yesterday, around potentially Blockchain disrupting ridesharing programs, right? Why do you need the intermediary if you have this open ledger and these secure transactions you can do back and forth with this ecosystem. So there's another interesting disruption. Now do the ridesharing guys proactively get into that and promote it, or do they almost in slow motion, get replaced by that at some point. So yeah I think it's a come-on on all of us, like you don't remain a market lead, every market leader gets destructive at some point, the key is, do you disrupt yourself and you remain the market leader, or do you let someone else disrupt you. And if you get disrupted, how quickly can you recover. >> Well you know, you talked to banking executives and they're all talking Blockchain. Blockchain is the future, Bitcoin was designed to disintermediate the bank, so they're many, many banks are embracing it and so it comes back to the data. So my question I have, the discussion I'd like to have is how organizations are valuing data. You can't put data as a value on, y'know an asset on your balance sheet. The accounting industry standards don't exist. They probably won't for decades. So how are companies, y'know crocking data value, is it limiting their ability to move toward a data driven economy, is it a limiting factor that they don't have a good way to value their data, and understand how to monetize it. >> So I have heard of cases where companies have but data on their balance sheet, it's not mainstream at this point, but I mean you've seen it sometimes, and even some bankruptcy proceedings, their industry that's being in a bankruptcy protection where they say "Hey, but this data asset "is really where the value is." >> Vellante: And it's certainly implicit in valuations. >> Correct, I mean you see bios all the time based on the actual data sets, so yeah that data set, they definitely treasure it, and they realize that a lot of their answers are within that data set. And they also I think, understand that they're is a lot of peeling the onion that goes on when you're starting to work through that data, right? You have your initial thoughts, then you correct something based on what the data told you to do, and then the new data comes in based on what your new experience is, and then all of a sudden you have, you see what your next friction point is. You continue to knock down these things, so it is also very iterative working with that data asset. But yeah, these companies are seeing it's very value when they collect the data, but the other thing is the signal of what's driving your business may not be in your data, more and more often it may be in market data that's out there. So you think about social media data, you think about weather data and being able to go and grab that information. I remember watching the show Millions, where they talk about the hedge fund guys running satellites over like Wal-mart parking lots to try to predict the redux for the quarter, right? Like, you're collecting all this data but it's out there. >> Or maybe the value is not so much in the data itself, but in what it enables you to develop as a derivative asset, meaning a statistical predictive model or machine learning model that shows the patterns that you can then drive into, recommendation engines, and your target marketing y'know applications. So you see any clients valuate, doing their valuation of data on those derivative assets? >> Altshuller: Yeah. >> In lieu of... >> In these new business models I see within corporations that have been around for decades, it's actual data offers that they make to maybe their ecosystem, their channel. "Here's data we have, here's how you interpret it, "we'll continue to collect it, we'll continue to curate it, "we'll make it available." And this is really what's driving your business. So yeah, data assets become something that, companies are figuring out how to monetize their data assets. >> Of course those derived assets will decay if those models of, for example machine learning models are not trained with fresh, y'know data from the sources. >> And if we're not testing for new variable too, right? Like if the variable was never in the model, you still have to have this discovery process, that's always going on the see what new variables might be out there, what new data set, right. Like if a new IOT sensor in the baseball stadium becomes available, maybe that one I talked about with elevation of the pitcher, like until you have that you can't use it, but once you have it you have to figure out how to use it. >> Alright lets bring it back to your business, what can I buy from you, what do sell, what are your products? >> Yeah so after being in business analytics is Cognos analytics, Watson analytics, Watts analytics for social media, and planning analytics. Cognos is the "what", what's going on in my business. Watts analytics is the "why", planning analytics is "what do we think is going to happen?". We're starting to do more and more smarter, what do we think's going to happen based on these predictive models instead of just guessing what's going to happen. And then social media really gets into this idea of trying to find the signal, the sentiment. Not just around your own brand, it could be a competitor recall, and what now the intent is of that customer, are they going to now start buying other products, or are they going to stick with the recall company. >> Vellante: Okay so the starting point of your business having Cognos, one of the largest acquisitions ever in IBM's history, and of course it was all about CFO's and reporting and Sarbanes-Oxley was a huge boom to that business, but as I was saying before it, it never really got us to that predictive era. So you're layering those predictive pieces on top. >> That's what you saw on stage. >> Yes, that's right, what, so we saw on stage, and then are you selling to the same constituencies? Or how is constituency that you sell to changing? >> Yeah, no it's actually the same. Well Cognos BI, historically was selling to IT, and Cognos Analytics is selling to the business. But if we take that leap forward then we're now in the market, we have been for a few years now at Cognos Analytics. Yeah, that capability we showed onstage where we talked about not only what's going on, why it's going on, what will happen next, and what we ought to do about it. We're selling that capability for them, the business user, the dashboard becomes like a piece of glass to them. And that glass is able to call services that they don't have to be proficient in, they just want to be able to use them. It calls the weather service, it calls the optimization service, it calls the machine learning data sign service, and it actually gives them information that's forward looking and highly accurate, so they love it, 'cause it's cool they haven't had anything like that before. >> Vellante: Alright Marc Altshuller, thanks very much for coming back on The Cube, it's great to see you. >> Thank you. >> "You can't measure heart" as we say in boston, but you better start measuring. Alright keep right there everybody, Jim and I will right back after this short break. This is The Cube, we're live from Fast Track Your Data in Munich. We'll be right back. (upbeat jingle) (thoughtful music)

>> Announcer: Live from Las Vegas, it's theCUBE covering InterConnect 2017 brought to you by IBM. >> Hey, welcome back everyone. We're live here in Las Vegas for IBM InterConnect 2017. This is theCUBE coverage of their cloud and big data event Watson Analytics, and IoT Cloud. It's theCUBE coverage for three days. A lot of great interviews. I'm John Furrier, my co-host Dave Vellante. Our next guest is Scott Francis, an entrepreneur, CEO, co-founder of BP3. Welcome to The Cube. >> Thank you, glad to be here. >> Great to have an entrepreneur on because you've been, in your business, you co-founded it, built it form the ground up, >> Scott: Right. >> Hundreds of employees. Now, over 100 employees. >> Scott: Right. >> IBM partner, great story. >> Yeah, we started with just two of us 10 years ago. And, we'll have our 10th anniversary in May this year. >> John: Congratulations. So take us through the, you know, state of the art. I mean, go back 10 years ago. You've actually provisioned your own servers. You actually had to load routers and networking gear. That's like, I'd say a tax of at least 100K in just gear. And then you've got the ISP chart, all that stuff. >> Right, well the economics have totally changed, right? For us and for our customers, and I think the main benefit is you can get to business value so much faster now and spend less money that's sort of wasted spend, right? >> So take a minute and talk about what you guys do and what your role is here. And then I want to get into some of the things that are changing the market place, that people are seizing opportunities around, certainly around processing and new innovations. So, give us a quick update on who you guys are, and your role here today. >> Yeah, so our focus is on business process and decision management. And, you know, our experience is that it is foundational technology and foundational aspect to almost everything you're hearing going on, right? Whether it's block chain or cognitive, or moving to the cloud. What are their key considerations? How does it impact my business process? How does it impact my operations? How does it impact my decisions? So we feel like in our space, we're right at the sweet spot of what all our customers are worried about. And when we hear them talk about block chain, we know we've got a process problem we've got to address. And when we hear about moving to the cloud, we better address all the Halo applications around that, application that's moving to the cloud and make sure they're all addressed and part of the new business process. >> It's interesting, the whole decoupling of existing systems models >> Right. >> Is really kind of what I see as the micro trend over the past six years, and like you mentioned, foundational building blocks is key, right? >> Scott: Right. So that's key. And, so let's take this to the next level. I want to ask you a question because I think this is something we see all the time on theCUBE when we do interviews, is that technology now is so much different. In the old days it was, we knew the process. >> Scott: Right. >> And we don't really know the technology. Let's go automate that accounting, blah, blah, blah. You know we saw that, ERPs, CRM, all those vendors. Now it's, I have technology, I don't know what the process is going to be because some new, big data analytics people changed the insight, and changed the value chain, or changed the business model, one tweak radically will disrupt proven, process which no one wants to change. Whoa, you know, so there's now a real factor. Give us some insight and color around how that goes down, because someone has an insight, they want to roll it in and implement it. It changes the entire process flow. >> Right, well the key thing is, having an insight as a single person in a process is one issue, but rolling it out across a Fortune 500 company is a whole other proposition, right? You've got regulatory issues and compliance issues, and customer experience issues that you've got to work through. And all those accommodations may be there. The value prop may be there, but you've got to work through it. You can't, you know, at a billion dollar organization, you can't just change it for that, you have to work all that out. >> John: So what's the playbook? >> Yeah, so the playbook is when we have an insight, what we talk to customers about is you've got all these tools now to arrive at insights you couldn't get to before, or by the time you got to them, you're doing your analytics over data that's six months old. Okay, now I have an insight about what would've worked six months ago. The difference is with cognitive and machine learning algorithms, and the analytics you have available today, and the access to the data, those insights are available now. We have to re-architect the processes to reflect that and to let me make new decisions within that operational context. >> Go ahead. >> Operationalizing those insights. Go ahead, finish your thought. >> Well the data first thing that you talked about is key. We just had our big data event. It's look in value in conjunction with strata hadoop was data in motion and badge are working together now to your point, the times series of data is relevant in the time you need it, right? >> Scott: Right. >> Not yesterday. So this brings up the question of, Okay, you've got some spark thing going on. I see IBM has got spark, that's cool. But now, how do you get into the app, right? To developers? I'm a developer. I'm a coder. Do I need to be a wrangler, data wrangler, or data scientist, to make that happen? So this is the conversation people are trying to figure out. What's your perspective on that? >> I think a lot of the tools that are, that are available now, basically made a common coder, right? Has a decent chance OF that competing with their data scientist friends. There's a different level of expertise, obviously, for the data scientist. But much like in business process, you know years ago, you had to get your lean six black belt, and you really had to study it to get good at it, and really master statistics, and I've got tools that will run the statistics for you, right? So you don't have to master the statistics but you've got to collect the right data, you have to engage in the business. So I think you see a sort of, democratization of data science, right? With the tools that are available now. >> Talk a little bit more about decision management. Go back to the mid-2000s and the Harvard Business Review is writing articles that gut feel trumps, you know, paralysis, analysis, paralysis by analysis every time. That's seemingly changed but what specifically has changed in regards to operationalizing those insights? >> Well I think they're a couple of things that are interesting. If you look at how processes were traditionally designed, you know, before BPM came along, BPM and decision management tools came along, just write the code. Build your application. And when you wanted to change the decision, well you had to find where that was modeled in the code, and edit the code, right? And that was a challenging proposition. The guys that wrote it might have moved to other projects. So how do you figure it out? >> So gut feel was faster. >> Yeah, and BPM, and OEM, you know, gave us tools for managing those things. BPM in terms of process, having a diagram that a mere mortal can understand and find the right context for whenever that decision gets made. And decision management to mange rule sets and the interactions between these rules in a more codified way that again, mere mortals can understand, right? So you don't have to go hunting through code. We're looking at a model, a representative model. I think the change now with machine learning, with cognitive computing, the real time access to data is that you have to really rethink your processes and allow those decisions to be altered in real time, not later, six months later, when I'm doing a revamp of the process as a separate, sort of institutional operation but actually as I'm running my process. We design it to accommodate the idea that as we're collecting data we're going to learn and get better, and actually affect those decisions, or recommend a different decision to the person whose Johnny-on-the-spot. >> Are you finding that the business impact is that your customers, the consumers of this sort of new way of doing decision management are seeing things that they wouldn't have seen before, or is it more greater conviction and faster time to everybody pulling the same direction? >> Well, I think for sure they're seeing things they haven't seen before. We're surfacing data that they just didn't have access to before in a timely fashion. And in the context of their process which was always a difficult thing to do in traditional systems, right? For any of your traditional ERP, or CRM system, the notion of where you are in your cross functional process may not be present. Today you have that context. You have the real time access to it. That really changes the nature of what you're seeing. I think the other bit is, yeah, the action ability, right? How easy it is to turn that insight into an action. >> And have you seen any effect on the politics of decision making, because we all know the P and L manager whose the strong voice in the organization, he or she is going to pull data that supports their business case. Have you been able to, sort of, neutralize that sometimes damaging effect in organizations? >> Yeah, well, I think in the cycle of the economic cycle, you know, if we rewind five or six years ago, almost every project we engage with with a customer is about operational controls, reducing costs, trying to produce the same result with fewer resources, right? And that has shifted dramatically over the last few years. The last two years it's been almost entirely about capturing revenue. >> Dave: Opportunistic, yeah. >> Serving new revenue streams without having to hire as much to support it. It's much more about revenue capture and customer experience. And I think that reflects the stage we're in in the cycle. >> Dave: Is that a bubbling cater? I hope it reflects a good long term view. >> Dave: I hope so too. >> You know, but it's interesting. There's a customer speaking here at InterConnect today, StubHub, about their customer experience. And they BPM to manage their customer experience, and back in 2009, 2010, when everybody was pulling back, and they were all focused on cost containment. You know, I recall StubHub was working on how to make their customer experience better. It's kind of interesting, right? And they've done very well over the years, right? So I think that value system in that culture really pays off over time, but you have to really mean it. If you're just swinging back and forth with the ebb and flow of the economy, then I think it's very difficult. >> Well, if you're doubling down when everybody else is sitting on their hands, you're going to get a competitive. >> It's a great opportunity, right? >> So, talk a little bit more about the IBM connection. What's going on in InterConnect, and what's the relationship there? >> Well, IBM is our best partner. You know, we've been partnered very closely with IBM ever since they acquired Lombardi which was our company that we came out of back in 2007. And that has become, you know, the heart of the IBM, BPM portfolio. And we work with their business process products, decision management, as well as cognitive and blue mix. So we're in the mix with IBM in a big way, and I think this conference is a great opportunity for us to not only reconnect with folks from IBM, but also with our customers who tend to come to this conference as well. So it's a great opportunity for us. >> So specifically you're leveraging IBM tooling, sort of. >> That's right. >> Repackaging that in your solutions for your clients. >> Right. So we are a reseller. We're also OEM IBM software, and we do delivery work for IBM customers. So, it's kind of a trifecta. >> You started this company 10 years ago. We love this start up story. Tell us, you and your colleagues started. Tell us your start up story and how you go to where you are now. >> Well we were, you know, we would meet up at a coffee shop, right? And get together and kind of talk about, you know, the fact that it felt like there was a big opportunity out there. >> Dave: This is in Austin. >> Yeah in Austin. My co-founder and I, you know, we were working at Lombardi but we felt like there was an opportunity to build a great services firm in our space, right? In this business process space, that there was a lot of untapped potential. And as we met and talked about it, we just got the bug that we needed to go out and do it. And when we started the company, you know. It was just the two of us initially. We bootstrapped the firm. Last summer, for the first time, we actually raised money, outside capital, to help fund the growth. >> Dave: 10 years then. >> Yeah, yeah. But all that time we self funded which was a great experience. A great learning experience. Certainly lost some sleep over the years. But, you know, there is an aspect of kind of putting the band back together. You know, hiring people we really enjoyed working with in previous lives, previous jobs, and putting together a killer team to go after it. >> So the decision to take outside capital, maybe talk a little bit about that because that's probably wasn't an easy one, or maybe it was, I don't know. >> No, I think, you know, what we've been fortunate to do is we've taken some calculated risks over time, right? We used to only operate in the United States. We acquired a business in London to expand to Europe. And now a third of our business is in Europe. But those risks, you can put the whole company at risk taking a chance like that. And so it occurred to us, after taking a few of those calculated risks and winning that maybe we should hedge our bets a little bit and have some more capital to work with, and have a good financial partner that if we were engaged in that kind of discussion, someone who could help, both advise and also possibly fund if we got into that situation. And so, we took an investment from Petra Capital based out of Nashville. They're a great growth equity firm, and they invest in healthcare and tech start ups, like ourselves. And so we got some great people on the board as a result. Mike Simmons from T2 Systems, and Jeff Rich from another capital investment firm. These guys have been operators. They've run companies much bigger than ours but they've also been in the mix at our size. So we've got some great outcomes out of taking that investment. >> So you've been cashflow positive since the early days. You had to be. Is it the plan to continue to do that, or do you make gasoline in the fire type investments? >> You know, I think it's cultural, right? I know there's a lot of business models where there's actually some good since in the running and not worrying about profit for awhile, but I also think you need to develop habits and our business serving enterprise customers, I think they deserve to know that we're being responsible with our money, with how we spend, with how we grow, and that we have a responsible level of growth. We could spend more and grow faster at the same type of process. >> John: At the risk of service. >> But at the risk of service quality for our customers and that's not worth it for us because ultimately, it's the repeat business with customers that really drives our growth long term. >> We feel the same way, obviously self funded. You know I'd say Silicon Valley is a story like that. Heirarchy of entrepreneurs and it's well known that the number one position is self funded growth without outside capital. It's a lot harder. No offense to my VC funded friends. It's a lot harder to do it from the ground up than just get other people's money. So tier one is do it yourself, which you guys are in. Get some capital, grow that and have an exit. Three, try and fail, or four, work for a company. (laughs) >> I think the key thing is it takes patience. If you're going to do it yourself and self fund it, you know, let the business fund itself, not just throw in your own personal money, but actually make the business fund itself. You have to have a lot of patience to stick with it. And I think whether by hook or crook, we picked a space that afforded us some of that patience, right? >> Yeah, you get rewarded for innovation. You get awarded for good service delivery. >> We feel like business is a human endeavor, right? So a good business process and good decisions are going to be problems that our children will face, not just us. >> And they're going to get more exciting for you as processes get automated with machine learning and AI right here on the doorstep, and Devops exploding with IoT coming on full line. It's going to change the game big time. >> Yeah, and I can't remember who said it but someone just yesterday was saying, you know, "It's not so much about automation "as it is about augmentation." And I really think that's true. I think if you automate out all the mundane, what's left is the stuff that's really interesting, right? And that's kind of how we view our job is to automate all the stuff that's getting in the way of highly skilled people doing their job taking care of their customers. >> I always love the story when IBM super computer beat Garry Kasparov at chess. You've heard this a million times. Kasparov didn't just say, "All right we're done." He created a competition, and he beat the computer, and now the greatest chess player in the world is a combination of human and machine. So it's that creativity, that common atoria factor that's drives the machine. >> It's actually better than the machine only, right? >> The creativity is going to change the game. Scott Francis, entrepreneur, founder, co-founder and CEO of BP3 in Austin. Thanks for joining us, appreciate it. More live coverage here. Stay with us, theCube is at IBM Interconnect here in Las Vegas. More great interviews after this short break. (upbeat techno music)

>> Narrator: Live from Las Vegas! It's the Cube covering Interconnect 2017, brought to you by IBM. >> Okay, welcome back everyone. We are live in Las Vegas at IBM Interconnect 2017, IBM's cloud and now data show. I'm John Furrier with my co-host Dave Vellante. This is the Cube. Our next guest is Derek Schoettle, the general manager of Watson Data Platform, and Adam Kocoloski who's the CTO of the Watson Data Platform. Guys, welcome to the Cube. Good to see you again Derek. Great to see you, welcome Adam! >> Thanks, John. >> So, obviously the data was a big part of the theme. You saw Chris Moody from Twitter up there, obviously, they have a ton of data. I like to joke about they have a really active user right now in the President of the United States. >> Daily State of the Union, I think, was the one take away. >> Daily State of the Union. But this is the conversation that's happening in all over IT, and enterprise, and cloud, both public and enterprise, is the data conversation in context to cloud. Super relevant right now, and there's architecturals at play, it's app, it impacts app developers, it impacts architectures. And that's the Holy Grail, the so-called app data layer or cloud data layer. What's your vision, guys, on this? Derek, I'll start with you, your vision on this data opportunity. How does IBM approach it? And what's different from, or could be different from the competitors? >> Yeah, I know, one, it's an exciting time. We were just chatting about before we went live is, there's so much change taking place in and around data, right? It used to be it's the natural currency, it's everything everyone is talking about. The reality is, it's changing business models, right? It introduces a whole new set of discussions when you introduce cloud, self-service and open source. So, when we step back and think about how we can differentiate, how we can make IBM's offer to clients and the broader market interesting, is shift to a platform strategy where it says, we have instead of discreet compossible services that act independent of one another that are not, I'll say, self-aware, shift into a platform where you have common governance, you have common management, and you have really a collaborative by design approach where data is at the epicenter. Data is what starts every conversation whether you're on the app dev side, whether you are a data scientist, someone who's, you know, at the edge of discovery. And cloud's what's enabling that, self-service is what's enabling that and operationalize is what we do. I mean, we spend our days thinking about and then operationalizing feature, function, and then performance for a lot of different workloads. 'Cause it used to be, I think the, I was at Vertica, right? So that was the introduction of volume, variety, and velocity, right? Now, with the introduction of AI and cognitive, it's really about taking any and all and rationalizing it. And any and all meaning sitting within your corporate structure, as well as what's more broadly in the internet, out available within social media, right? That to me is the shift that's taking place. It's all companies are realizing they made a lot of investments, they have a lot of data, and they're not taking advantage of it. And we see that the big shift is... People are saying data scientist, what we think about is the merging of data and science. You think of science as cognitive and AI, right? That's a small population that really understands and can take advantage of. You have a whole big market that's out there in traditional data and analytics. Our platform is about merging those two. It's really about merging those experiences so everyone takes advantage of the benefits of data and science. >> What's the conversations that you are having, Derek, with customers? Because I think that's, there's a lot of bells going off into the CXO or even practitioners when you hear about machine learning, you hear AI, cognitive, autonomous vehicles, sensor networks. Obviously that's, the alarms are going off, like, I'd better get my act together. So, how do they pull that off? How do your customers pull off making that happen? Because now you got to bring in to be cloud ready, you have all these decoupled component parts. >> Yeah. >> John: You got to operate them in the cloud and you got to kind of have an on-prem component that's hybrid. What are the conversations that you are having with customers in how they're pulling this off? >> Yeah so, I'll cover the first piece, and I know Adam is spending certainly this week and a lot of time as well with clients on this topic. You know, the first part of the discussion is do you believe that the cloud can help you? Most folks are saying, "Yes, we believe it can help". Second piece is, how do I take advantage of emerging technologies that are moving at a rate and pace that perhaps my skills, my existing IT architecture, and my business model can't fully kind of, grasp, if not take advantage of? So, what we've introduced is a methodology, a data first method, which literally is a, it sounds simple, but at the end of the day, it is a common, uniform, agile way for us as IBM to engage with partners and clients that literally starts with the discovery workshop that says how does data inform your business? It's not static reporting anymore, it's what is the data that's sitting within your organization? You heard it from James at PlayFab. Data is changing the way people build in games today, thinking about how to enrich games, so on and so forth. Data First Method is what we've introduced, so you'll see going forward, IBM will sell Data First, we will engage Data First. So, any conversation with someone who says, "How do I take advantage of AI, "or machine learning, "or data science experience?". Well, let's step back for a second and talk about data. 'Cause 30 years ago, 20, that's how every conversation started. You get on a whiteboard, you design a schema, you talk about the relationships. That's how it started, and we're kind of cycling back to that, right? We got to put data first. >> So, Adam, the geeks are always arguing speeds, "I got a Hadoop cluster here, "I got this over here.". I mean, there's a lot of variety and diversity in terms of how people can manage either databases, and middleware or what not, right? So, how do you see the data first? How does it play out architecturally? And how does that play out for the solution? >> I think one of the big advantages we have in the world of the cloud platform is this opportunity to, on the one hand, use more a broader variety of compossible services, but also be able to take different parts of the business that were historically a little bit more separated from one another and bring them together. So you look at a Hadoop-flavored data leg on premises. It's a good area to do discovery, a good area to do exploration. But what clients really care about time and time again, a common refrain is the operationalization of the analytics, of the machine learning models. How do I take this insight that my data science team has discovered, and have it really influence a business process or incorporate it into an application? And in the on-premises architecture, that's often times quite a challenge. In the world of the cloud platform and the Watson data platform, we have an opportunity to be a little bit closer to things like the world of kubernetes which are really ideally suited for deploying and scaling microservices and APIs in a cloud-native, fault-tolerant, reliable fashion, right? So, you're seeing us take that menu of composable services in the cloud platform, and treat the data platform as one such composition. An opinionated way to put together this menu of services specifically to help data professionals collaborate, and drive the business forward. >> So, when you guys announced the Watson Data Platform, I think you called it Data Works, then changed the name, about five, maybe six months ago you messaged that 80% of, you know, data professionals' time is spent wrangling data, not enough time doing the fun stuff. And the premise was you coming up with a platform for collaboration that sort of integrates those different roles as well as, as you pointed out just now, allows you to operationalize analytics. Okay, so we're five months in, six months in, what kind of proof points do you have? Have you seen it? I mean, some people were skeptical saying, "Okay, well, it's IBM, "they've put a nice wrapper on this thing, "pulling in some different legacy components, "and you know, nice name." Okay, so, what do you say to that? And what evidence do you have that what you said is going to come true is actually coming true? >> You're going to do tech and I can do customer? >> Yeah, go for customer first. >> Yeah, so what we've seen is if you think about why we ended up at a platform. So, if you roll the tape back to when Cloudant got acquired in 2014, the journey that we were on was everyone was building rich applications, they wanted to be smarter, they wanted to understand what that exhaust was coming off. >> Right. >> Derek: And they wanted to add different ingredients to it. So, instead of a do-it-yourself kit that is a bunch of proprietary interoperability issues that's a ton of expense and inefficiency, and can't take advantage of the cloud, we decided, in very much of then our path towards, let's build a platform that allows you to easily ingest, govern, curate, and then, I'll say present and deploy. So, starting in actually June, and thhis started first with Spark. We made a huge bet on Spark 'cause we believed that to be kind of the operational operating system, if you will, for an analytic fabric. So, it started in Spark. Then, when we announced the Watson Data Platform in October it was, here's how we're going to take our heritage run governance, our heritage run traditional structured, non-structured data repositories, and here's how we're going to take visualization and distribution of data. So, that then next went into how we bring it to market? That's Data First. So, we've been working with large insurance companies, large financial services companies, retailers, gaming companies, and the net that we see is three things. First is, yes everyone agrees the platform is the right place to go. It's where do we get started? How do I take my existing investment and take advantage of this platform? And that, invariably, is I'm going to build a net new application whether it be Watson Conversations, so that runs into Watson Data Platform. We want to ingest data, but we want that data to be resident on-prem, we want it to be native to the cloud, and so we're going to work through the architectural change to adopt that. Another great example is we want to start with just an analytic application because we are already hosting with you a mobile app. Well, we're going to run it into your analytic fabric using dashDB, and dashDB works with Watson Analytics and we're going to build an application that's resident. The really creative and compelling piece here, back to your comment on IBM is, it's really hard to buy things from this company historically. Buying things from IBM is not easy, so we built a platform, we built the methodology to help you understand how to take advantage of it, and now we have a subscription, the Bluemix subscription is which you can come in and draw down those services, be it an object store, be it a sequel data store, be the visualization layer. >> John: Opposability basically. >> Yeah, but in a common governed framework. The big takeaway is, and I'll pass to Adam, governance and security and operationalizing the platform is what we can bring to bear. 'Cause we're bringing Open Source, we're bringing proprietary technologies, but if it's done independent, it doesn't really deliver on the promise of a platform. >> I will say that architecturally, that's incredibly liberating to know that there is this one common mind model. >> It's also highly requested by customers. That's what they want. >> Derek: That's what they want. It's the path to get there that I think is, we're at that intersection right now, it's crossing the chasm. >> John: So, what's liberating? Give us good-- >> Oh, just the fact that you know that if there's a common access control layer under the hood, if there's a common governance layer under the hood, that you don't have to compromise and come up with an alternative proposition for taking some capability, maybe deploying a model to a scoring engine. You can have the one purpose filled scoring engine and know that I can call that in on demand from discovery phase to go to production and I don't have to sort of engage in another separate mind conversation or separate entitlement conversation or a separate enabling conversation. This catalog is allowing it to work together. >> That to me from a team sport perspective is that the steps you have to take. So, think of ETL. ETL really in a modern real time, like getting away from batch and go into real time, that's just flow. So, the skill set and the ownership of the infrastructure associated with that is evolved, especially in cloud where that's just a dynamic where it's going to be a team deciding here's the data I want, here's how I want to enrich it, here's how I want to govern and curate it. >> It's a team sport. I love that. We were just at the Strata Hadoop. We had our big data SV event and the collision between batch and real time, they are not mutually exclusive and some people just made bets on batch and forgot real time. And they have real time people who don't do batch. So, you kind of see that coming together. >> Adam: Conversion. >> So, the question, Adam, for you is that, with the world kind of moving in that direction, how do you rationalize so the customer who's saying, "Hey, I'm cloud native but I also have a hybrid here "and I want to be cloud native purely "on this net new applications". So, there's a conversation happening. I call it the dev ops of data which is like data ops. Hey, I'm a programmer. I just want data as code. I just don't want to get in the weeds of setting up a data warehouse, and prepping an ETL, all that batch stuff that someone else does. I'm writing some software. I want data native to my app, but I don't want to go in and do the wrangling. I don't want to go out. I just want stuff to magically work. How do you tackle that premise? >> I mean, I think the dev ops of data piece is certainly a topic we're going to be hearing a lot more about over the next coming six months, in a year. I think the reason for that is precisely because this earlier topic of operationalization. You've got lots of people building up, budding data science teams and so on. And the first thing they're going to do is be working in the discovery area. They won't be in the world of pushing things to production. When they do, it's going to become more important that the folks who truly understand the details of the algorithm are close enough to the deployed assets, so that they can understand how this model is behaving over time. So that they can understand new data quality issues that might have cropped up and get close to that without obviously sort of breaking the separation duties that are important for a production system. So, I think, that is one part of the data ops conversation that hasn't yet been worked out. It's going to be a real opportunity for folks who-- >> That's an emerging area. You agree, right? >> It's a cultural shift too. I mean that is a re-thinking of, because most companies keep data in steel pipes. They're highly regulated. Their rules, the personalities that own them so to speak. The proposition that we've been on and every client asks for is how do I create a common fabric that gives access to people, that is governed and curated so you can always give a shopping experience. People that work with data do not want to talk about and say this : "How long does it take to stand up a server? "When can I get the data stood up in the staging area "so I can actually access it?" That's over. >> It's interesting, we're doing some Wikibon research on this, and this is the point where people look at value extraction of the data so they tend to, it's kind of like if you're a hammer, everything looks like a nail. So if you're in IT, it's infrastructure. If you are on the business line, it's the apps. So, you're seeing the shift where apps is value creating the value, but the infrastructure is more elastic, more compossible so it's enablement by itself so that's interesting. So, your thoughts on that, guys? Where is that value of the data coming from most, right now? Is it the apps? Is the infrastructure still evolving? The hybrid not-- >> We think there's a value model here. There is certainly elements of the data pipeline that are purely operational, reporting base and things like that, which drive value on their own. But we also recognize that it's new uses of data and new business processes that are primarily driven by applications, driven by conversational interfaces, driven by these sort of emerging paradigms. And one of our goals in the data platform is to ensure that clients can move along that curve more aggressively. >> How are people getting started with the Watson Data Platform? Do they go jumping all in? Is there a community edition, you can try it before you buy it kind of thing? >> Yeah, so you're signing up in Bluemix. You have access to a set of services around the platform. You have a 30-day window where you can try everything included within it, and then at some point you got to commit to a credit card or you got to commit a 12-month term agreement. I think in parallel, we see a lot of other companies that end up blasting in size challenge for IBM. We have a lot of clients. We have got a lot of clients that we are working with today in traditional architects and infrastructure, helping them through a methodology, helping them with the right skills. That is a more traditional, hey, come in and try an analytic workload on the platform. We'll give the skills. We'll help do the enablement and then we're off and running. I think the big difference is whether or not clients are paying for and they are willing to pay for it. 'Cause we are helping them get to this new model. We're helping them get to the platform, and I think the big thing we're working through is how do we get to velocity? I think when you look at these workloads that are happening. The reason they're happening is now data is not just in some dark corner. With AI, the machine learning is always on. So, there's a lot of different ways in which you can unleash that, that then, how do you take advantage of it? And that is a cultural shift. It's re-thinking business models, it's re-thinking how you got skills deployed which is incredibly exciting for us, and I think the market in general. I think back to how AI is cast in many cases as the robots are going to rule the world. There's a lot of good that can come from exposing vast amounts of data to AI and to frameworks where you can get a lot of value out of it. From how to better position products to how to, better design of medicines to fulfillment chains in countries that need help. >> So, guys, in the last minute that we have I want you to take a minute to either together or one of you guys talk about how IBM is helping solve what seems to be the number one question we get on the Cube where I get asked, hey, how do you help me build a hybrid architecture. I have more data-rich workloads coming on board now. Either I have some heavy data rich workloads that are run on-prem, I got more cloud action coming, I got IOT and I'm investing in data science. So, how do you guys specifically help me build a hybrid cloud architecture that's going to fuel and support data-rich workloads and propel my data science operation. >> Yeah, so, I'll take the basics for me. It is the Data First method. It is dashDB, which is an extensible on-prem hybrid in the cloud so that the common analytic fabric. There's Data Connect, which is our ability to move data batch continuous into different end states in the cloud, and then there's data science experience. So data science experience is our offering that brings together community, it brings together content, it brings together various tooling for the data scientist or data engineers. And I think the other piece of this is, we have something called solutions assurance. So we're literally designing patterns that we stand up in our own environments that reflect what we see on Premise and what we see workloads going into the cloud with, and stamping that as hybrid architectures that are repeatable, and we remove risk, the operational risk. But the reality is (mumbles) is, clients have to make sacrifices in getting to the cloud. You have to deprecate, you have to rethink. And that's where some of the smoothing of those rough edges come into the discipline of us saying, here's a supported architecture, here's the destination that you're going to, and we're going to have to work together to get there. Which is the fun part, I mean, that's what we're all in this for, is getting the outcomes. >> I think the key is not to pretend that these environments are completely identical to one another. There are things that the public cloud is uniquely well suited for. So let's make sure that those kinds of use cases are really nailed there, right? And then there are other cases where you're dealing with mainframe systems running critical business processes, and you want to be able to infuse that process with some analytics. So you have to look at the use case. Maybe it's training a machine learning model in the cloud, being able to export that model and run it-- >> So use proven solutions and be prepared to be handling new ones coming onboard. Alright, Derek Schoettle, general manager, and Adam Kocoloski, the CTO, the leaders at IBM Watson Data Group, IMB Watson Platform. This is The Cube, back with more live coverage after this short break.

>>Hi from Las Vegas, extracting the signal from the noise. It's the cube covering IBM insight 2015 brought to you by IBM. Now your host, Dave Vellante and Paul Gillin. >>Welcome back to IBM insight everybody. This is the cube. The cube goes out to the events. We extract the signal from the noise. This is I think our fourth year at IBM insight, IBM's big, big data show. IBM doesn't use that term, they call it analytics and it's been done a tremendous job of taking this giant portfolio and then building a leading the leading actually analytics business in the industry. Harriet Fryman is here, she's the vice president of marketing at IBM analytics. Harriet, welcome to the cube. Good to see you. Thanks for having me back. Yes, so the show here is big, I think bigger than anyone, you know, we've been to a lot of great energy. The solutions expo is tremendous. The, the keynote this morning were packed the general session, so you must be thrilled. >>Yeah, it's fantastic audience. And we just came off our advanced analytics keynote this afternoon. We were talking about the advances in Watson analytics. So the smart data discovery tool as well as our new release of Cognos. >>So Watson analytics is just permeating all parts of the business in the healthcare business, the cloud business, the analytics business. Talk about the impact that that little sort of experimental program with jeopardy has had on the company as a whole. >>Yeah, it's really delivering on the promise of, we talk about around the cognitive business and where Watson analytics comes in. It's really looking to bring that smart data discovery to an individual on their, um, on their PC to get instant insights into data. Whereas before they're really, um, could get access to the data, but how do they find the causation between data points versus just take a look at sales data, finance data. So Watson analytics really allows them to have that natural language question and um, have the processing behind the scenes find the interesting stuff in the data. >>Big idea is a, is it a marketing executive? You've got to love the, the fact that you can actually produce such a capability, you know, it's not like a little point product that's a platform that can touch every part of your business that can change lives. What are your, can you comment on that as again, from a marketing perspective, >>it's always fun in marketing to have a great portfolio to be able to market and something that really makes a difference to people's business. So with the, with Watson analytics and with what we're doing with Cognos around our business intelligence, it's great to market. Um, what has always been promised, I think in the BI market for many years, which is self service analytics for all. So, uh, as we're marketing both the capabilities around Watson as well as the capabilities and Cognos, it's kind of a delight to say, you know, what we were talking about give insight to everybody to make better decisions. It's really coming to fruition. >>If IBM has grown its analytics business largely through acquisition, I think you'd have 25 acquisitions. You've got a of different great brands, SPSS, core metrics, Cognos and the like is Watson, they're going to evolve and do a kind of a simulation point for all of those? >>Well, yeah. What we look at is, um, as we talked about the cognitive business and Watson really been the cognitive computing engines of, of that business. We're looking at how our analytics business really expands a company's business companies, company's ability to really understand what the data is, turning them, learn from experience of working with the data and put that into practice. So we can do that with dashboards, with reports as well, which is help people understand there's insight to be gained from data. There's value to be gained from data. And so you can apply it through being a learning company with or without having a cognitive system itself. It's, I'm going to take data, I'm going to apply analytics to understand patterns and I'm going to apply that to my business. And then I'm going to learn from the feedback loop and just keep learning, learning, learning. And that's what a cognitive business is about. >>So the BI business historically, you know, it's been interesting to watch. I mean I remember when it was called, you know, decision support, right? And, and it's put on a lot of promises, 360 degree view of the business, you know, predictive analytics and it didn't live up to those promises. And then you have this whole Hadoop movement come in and they're going to live up to those promises and then you realize, wow, they actually can't live up to those promises without the traditional data sets. And are those two worlds coming together? Is that the way that we should be thinking about this to actually fulfill on those promises? The last 15 to 20 years? >>Yeah, I think we always had the chicken and the egg, right? You can't have great analytics without great data. And what's the use of great data and as you have great analytics, so you really need both together. And then the promise has always been a great three 60 degree view of customer actually requires being able to get your arms around the data itself, reconcile it, make sense of it. And then it requires great analytics and a way to deliver it to the people who can use it in their business, be they in call center and service and sales. So the promise has always been there is the fact that we need to put it all together. We need to put together the data, as you said, Hadoop and relational data altogether inside and outside the firewall. We need able to make sense of it. So bring those entities together, do master data management, make the data, make sense as you pull it together and then have a great way for people to understand it. Consumer apply it in their business. >>So Cognos was obviously huge acquisition don't, Paul wasn't mentioning many of them. I think we used to tell you it's one of his favorite and I think it was rather large. It was with $5 billion acquisition, I believe. And so, and then IBM has sort of supercharged that entire business. So how has Cognos evolved and where are we today? >>Yeah. So as, as I came in through the Cognos acquisition many years ago when IBM acquired us, I really have seen it just develop and expand from the day that we, uh, we came on board with IBM. It's really expanded in a couple of ways. One is that we have expanded, um, cognitive capability to get at all types of data. So you mentioned Hadoop. So now we've, we know that in order to deliver a rich understanding of what's happening in the business called the Cognos reporting capabilities need to access all of that data. And so it does, it can access relational data, data and appliances, Hadoop data, data on the cloud. So really expanding the Corpus of data that can be put into a report and consumed by business. The second, a big investment has been, um, where BI was always thought of as an it only tool. Now I ask it for a report. They have a report backlog. Some months later they may give me a report. It's not quite what I wanted. That whole world has changed now, which is really bringing BI, we imagined into business people's hands because they want the right to be able to model data to be able to author reports, distributed, shared among their colleagues. So it's been an exciting journey as we've really taken business intelligence really to the next level. >>It's all about the, the role. What's the role of the spark, the big spark initiative that IBM announced a couple of months ago vis-a-vis all of the analytics products, the spark act as kind of a preprocessor for the, the capable of the value of those, uh, those point products add or how does spark fit in with them? >>Yeah, so, um, so with our spark investment, we announced our commitment to spark back in June and since then we're really looking at as well what we coined the term, the analytics operating system. So we see it as that foundational layer that's really going to speed up the speed of analytics as well as be able to apply algorithms to a much bigger, um, Corpus of data than you traditionally would have in a statistics tool, for example. So since then, actually today we announced that we now have 15 solutions built on spark across our analytics and our commerce portfolio. A great example is we replatformed DataWorks, which is our ability for business to do data wrangling as part of the Watson analytics work process. So we see spark is really an enabling technology for ourselves and then we've committed a significant investment back into the spark community to keep it enhancing the core fundamental capabilities of spark so that everybody in the ecosystem can take advantage of that. >>He said something just a minute ago, VI re-imagined. I want to pick up on that theme because again, the BI world used to be insights for a few and then they were very productive, very productive few, right? They had a huge impact potentially on the company. But you now hear things like we heard this morning about you know, citizens and analytics and the likes. So, and you have the, you know, the BI for Hadoop vendor does your sort of attacking the old, you got the vis guys attacking that business. As we said before, it's still critical. But so what is BI re-imagined? You know mean that means more agile. It means simpler, it means embedded into the workflow or the organization. I wonder if you could describe that in some more detail. >>Of course. So when we look at business intelligence, I totally agree with you. It's really a tool that it use to develop reports or dashboards that were then delivered to the corner office, the suite for them to understand how my sales trending, what are my financials looking like, what's my production yield sort of reporting like. And that's great. Um, but that's kind of left a, a population that was not served, which was really the, uh, the business users who wanted to find insights for themselves. And that's really where the desktop discovery tools kind of were born, which was to satisfy that need out there that was not being satisfied by BI. When we're looking at re-imagining BI, we're looking at serving that community too, which means we have re redesigned the user experience of business intelligence so that those people out in the business can author their own insights, can distributed, distribute their own insights. >>And we've taken the learnings of how we designed Watson analytics and that user experience into the BI portfolio too. So let me give you an example. So for example, um, I'm looking for data. I want to report sales by product and by region. Um, I would have had to in the past have it build a model for me of that data. Now with re-imagined BI, I can be in the business, I can simply type in sales product, region. It's got to propose the data. So I don't need to know where the data's stored. It could be in Hadoop, it could be in relational. It's going to propose what data might be the most relevant to me. I can hit hit a button that says proposed model. It's going to model it for me in a way I go. So I didn't need to be a data modeler. I didn't need to know where the data was stored. So now I'm much more empowered as a business person. I don't need to offload that data into a desktop tool, worry about data silos, fragmentation of the decision process. I've now bought to that underserved population. >>So you've said what you've described, you've got a library of models and the system chooses the right one and fits for me. Is that right? Did I, >>you actually have a light. Yeah, close. You actually have a library of data sources and then you can build different models across those data sources. So you mentioned that there's a, a, uh, a dashboard tool right over here for Hadoop over here for maybe if another file system, etc. Well, that's great if all your data sits there. What we've done with BIS, we said, let's make that invisible and then you can pick data from any data source and bring it together into a single report. >>We had a routine of gunner on this morning talking about, uh, talking about governance. And what you're talking about was sort of democratization of, of analytics and, and everybody having their own, uh, their own tools, ability to manipulate data, I mean that has to proceed from a solid foundation of data governance. How well prepared our clients in your experience to proceed in that direction you're talking about they have that data really well hardened and bullets. >>So there's, there's a couple of steps I believe that um, clients understand that there's need to have integration and governance over the data sets, the challenges, the kind of Maverick use of data that happens in a company. So it's both tooling and technology as well as a corporate culture of how you're going to treat the data that you have in your, in your company. So where Ritika talks about the fact that you need to have a data reservoir, you need to have data warehouse, you need to have governance over that. We also need all of that governance to go all the way through to the end consumption of data. So where we've re imagined BI is to say you need that trusted source. It may sit on a server or many servers and need to make that available to everybody to self-serve and their first call to be, I shouldn't be, can I download that data into a tool myself? Cause the minute you cut that cord, your governance is gone. Now clients are starting to understand that because they're hitting that as the data discovery tools, um, start getting hold in the business, which is there's as many copies of data as people in the organization. And so one way to tackle that is to say no, I need to bring them back into the fold on the govern data and do that in a way that doesn't compromise their self service. >>So the big data meme sort of exploded around 2012 my, at the time, my 13 year old would joke and say good morning Polara and she'd say, morning daddy hashtag big data. And so I remember in 2012 when we came to insight, it was interesting to observe, but what IBM had done with this sort of bespoke portfolio of assets is put them together. And I said at the time, super glued it to the big data meme, changed the language around analytics and business outcomes and is now dominating that business or will dominate that business was kind of my prediction and it's exactly what you did in my, my version. Um, so let's talk about your portfolio. You've got purview over, so there's information management that's BI, the predictive analytics database is, is in there as well, and data integration, is that right? So there's that. What were once sort of these bespoke toolings talk about how you bring those together and bring them to market and message them? >>Yeah, yeah. It feels like there was, um, an evolution that happened in the marketplace, which is, as you said, it was almost like it had a shopping list. I'm going to go shop for BI now. I'm going to go and shop for predictive analytics and I'm going to go shop for a database and I'm going to go shop for integration. And really that's, um, great to have capability coverage. But in order to actually get insight from data, you need to be able to be in all the types of data, wherever it resides. You need to be able to put that data into context, which requires integration, master data management, and then you need to be able to deliver that, that, um, analytics and insight capability to everybody who needs it both through a dashboard as well as embedded into applications. So we really saw the opportunity to help our clients get value was to put them together and integrate them in such a way that you actually look for what business questions you want to answer. You don't shop by capability anymore. So the great thing when we look at how we market that is we can start with the business outcome or the client value and work back from there because different types of business problems require different combinations of the capabilities. >>And, and you find, I, you know, there's an old saying it's better to have overlaps than, than gaps. Do you find that you have more overlaps than gaps or do you find that you still got big gaps that you need to fill? >>Um, I think the language, we need some more English words and we need more words in the English language because when we say I need to get it data, I need to integrate it together and I need to deliver it. You could say that about Hadoop, right? Cause it does that. You could say that about a relational database. You could say that about our business intelligence tools. So sometimes people get, it appears like there's overlap because there's only so many limited words that we have to describe what we do. But it's the use cases that will prescribe which part of the portfolio we use. >>So at the, at the strata Hadoop world show this year, there were three or four big themes that emerge. You know, one was really about the data in motion in real time. You know, we talked about spark earlier. Uh, the second was the data, the database, the file system, you know, that sort of plumbing. Um, and the third was sort of complexity. Uh, everybody sort of choking on Hadoop complexity, spark helps but sparks complex too. So it seems like you guys are trying to take all that stuff and just make it invisible. Um, start with the business outcome and say, okay, you need real time. We, you know, to service this business or crime fraud, you know, is going to require some real time nature or maybe it's micro batching and whatever technology you use. Um, is that the right way to think about it that you're trying to hide that complexity and how do you hide that complexity? >>Yeah, exactly. We um, if you take the analogy of a car, everybody drives a car, but we don't necessarily have to understand how the engine works and you know, when we buy the car, we don't open up the hood and take a look and have everybody explain every single piece part and how they all work together. And that's sort of our destiny for what we're doing with insight, what we're doing with the solutions we build, which is yes, it has all those capabilities inside it, but you don't have to be technically savvy enough to understand what that is. You just need to know that it does what you want it to do for your business. So our is with data management, the hide, all the complexity of different data containers behind the scenes using big sequel or ways to access and make that transparent. Then with the analytics, we're looking to make the analytics transparent. So whether you're using an algorithm written in spark, you use an algorithm written in R, it doesn't matter. You're looking to have an algorithm apply to, to find patterns. >>But the way you would hide that complexity over the last 15 years is a big services engagement. And that's changing. Am I, am I understanding that right? I mean you're, you're changing that. You're driving more software into the platform and you're doing it with API APIs and, and, and less of an emphasis on leading with services, more of an emphasis on leading with business outcomes. And then mapping the technology to that. Is that, is that fair or is it still very heavily services led? >>Yeah, we definitely live the lead with the business outcomes. Um, as we look to support hybrid cloud environments, some of that technical complexity is, is made invisible because of the way that we use cloud. So you don't have to worry about deployment and enter production. The other thing we do with our services is we're much more focused on how are you going to apply the data that you have. How you get to apply analytics to actually change your business or services is much more in discussion of how are you going to make this impactful for your business versus the bits and bites of how do you install it, configure it and deploy it. >>But who, who is, who on the back end is going to do that dirty work. And who do you see in the companies you work with? Is there a specialized data function emerging within the CEO's organization? Is it, is it independent? Is it a set of independent of it is too important to the business or who who, who do you recommend do that backend plumbing work? >>Because we always used to talk about two populations in a client business and then it and how business and it would work together. We actually see a third leg of the stool happening, which is around the data professionals, so that's all the way from a chief data officer to achieve data scientists, data engineers, to application developers to implement those insights. So we see this third profession emerging in our clients. Now what's interesting is when they report into the it organization, they're more centered on data management, integration, governance. When they report into the business, they're much more focused on applying analytics for business outcomes, but you're absolutely right. There's this third data savvy PR profession that's really rising in importance and you see a lot more appetite in clients to get that data savviness as a population in the company. >>At this point, you don't see any pattern emerging for where that function lives in the organization. Does that so? >>Correct. We see two, two distinct patterns in it. To better manage the data in the business to better drive an outcome from analytics. >>Do you see this, is the CDO a coming role? Is that, is that a high growth function within the big corporations you work with? >>It's definitely a function that is pretty much becoming established. They're called chief data officers or chief analytics officers sitting at the table helping with the business strategy of how to apply data for a difference in. >>And is that something CIO should worry about? >>Um, I don't, I don't know if they were, I'd have to ask a CIO that question, but definitely the CIO world is shifting much more to how do I provide the it infrastructure as a service provider. And then the CDO is C D O is taking that role with the data and analytics. We'll wait to see how it falls. >>Well, one of the, one of the sort of sea level question I think was about two years ago, the garden forecast, the chief marketing officers would spend more than CEOs by 2017 on it. Are you seeing that really happen? >>We're definitely seeing that. Um, the business side, the CMOs, the VP of sales, the chief operations officers driving much more of the decisions around analytics and data. The other thing that we're seeing is, um, and I think IDC actually quoted this is the rise of the profession of data science. It's outpacing the rise of it. >>Yeah. I mean in terms of growth rate we presume interesting or Harriet really appreciate you coming on the cube. We gotta we gotta leave it there. But last question is sort of, when you think about insight 2015, think about all the, the developments that have occurred over the last say four or five years. So how would you sort of summarize where we are today? What's the bumper sticker on insight 2015 >>the bumper sticker on insight 2015 is as its name in first insights to outcomes. You talked about big data five years ago. We're really shifting from being data hoarders and worrying about what the, how much data we have and what type it is to being insight hunters, which is how can I get the insights I need to make a difference to the, >>and that's where the business value is. Harry, thanks very much for coming on the queue. It's great to see you. All right, keep right there, buddy. We'll be back with our next guest right after this. This is the cube. We're live from insight 2015 in Las Vegas. We'll be right back.