>> As we enter the new reality of hybrid work, the exposures that companies have faced as a result of the pandemic and consequent shifts in technology strategies, demand new ways to protect data. We heard today from INFINIDAT and one of its key partners, Mark Three Systems, how these two companies are helping customers respond to this threat. Key factors to consider are how to back up data as quickly as possible. And at the same time, isolate critical data in the case where hackers are attempting to hold your data hostage. Now in that instance, it's critical to have an isolated, safe copy of the data because as the saying goes, backup is one thing, but fast recovery is everything. This is Dave Vellante for the Cube. Thanks for watching. >> Keep rolling, Alex. Do one more day just as a safe copy. >> You got it. >> A little more, little more energy in your face, your hands, your pits. >> You bet. I don't know. You bet. Okay. Ready, Alex? As we enter the new reality of hybrid work, the exposures that companies have faced as a result of the pandemic and their consequent shift in technology strategies and spending it demand new ways to protect data. Now we heard today from INFINIDAT and one of its key partners, Mark Three Systems, how these two companies are coming together to help customers respond to new threats. Now key factors that customers should consider are how to backup data as quickly as possible. And at the same time isolate critical data in the case where hackers are attempting to hold your data hostage. In that instance it's critical to have an isolated saved copy of the data because as the saying goes, backup is one thing, but fast recovery, that's everything. This is Dave Vellante for the Cube. Thanks for watching.

from the cube studios in Palo Alto in Boston connecting with thought leaders all around the world this is a cube conversation hi buddy welcome to this special noob digital event where we're focusing in on data ops data ops in Acton with generous support from friends at IBM let me set up the situation here there's a real problem going on in the industry and that's that people are not getting the most out of their data data is plentiful but insights perhaps aren't what's the reason for that well it's really a pretty complicated situation for a lot of organizations there's data silos there's challenges with skill sets and lack of skills there's tons of tools out there sort of a tools brief the data pipeline is not automated the business lines oftentimes don't feel as though they own the data so that creates some real concerns around data quality and a lot of finger-point quality the opportunity here is to really operationalize the data pipeline and infuse AI into that equation and really attack their cost-cutting and revenue generation opportunities that are there in front of you think about this virtually every application this decade is going to be infused with AI if it's not it's not going to be competitive and so we have organized a panel of great practitioners to really dig in to these issues first I want to introduce Victoria Stassi with who's an industry expert in a top at Northwestern you two'll very great to see you again thanks for coming on excellent nice to see you as well and Caitlin Alfre is the director of AI a vai accelerator and also part of the peak data officers organization at IBM who has actually eaten some of it his own practice what a creep let me say it that way Caitlin great to see you again and Steve Lewis good to see you again see vice president director of management associated a bank and Thompson thanks for coming on thanks Dave make speaker alright guys so you heard my authority with in terms of operationalizing getting the most insight hey data is wonderful insights aren't but getting insight in real time is critical in this decade each of you is a sense as to where you are on that journey or Victoria your taste because you're brand new to Northwestern Mutual but you have a lot of deep expertise in in health care and manufacturing financial services but where you see just the general industry climate and we'll talk about the journeys that you are on both personally and professionally so it's all fair sure I think right now right again just me going is you need to have speech insight right so as I experienced going through many organizations are all facing the same challenges today and a lot of those pounds is hard where do my to live is my data trust meaning has a bank curated has been Clinton's visit qualified has a big a lot of that is ready what we see often happen is businesses right they know their KPIs they know their business metrics but they can't find where that data Linda Barragan asked there's abundant data disparity all over the place but it is replicated because it's not well managed it's a lot of what governance in the platform of pools that governance to speak right offer fact it organizations pay is just that piece of it I can tell you where data is I can tell you what's trusted that when you can quickly access information and bring back answers to business questions that is one answer not many answers leaving the business to question what's the right path right which is the correct answer which which way do I go at the executive level that's the biggest challenge where we want the industry to go moving forward right is one breaking that down along that information to be published quickly and to an emailing data virtualization a lot of what you see today is most businesses right it takes time to build out large warehouses at an enterprise level we need to pivot quicker so a lot of what businesses are doing is we're leaning them towards taking advantage of data virtualization allowing them to connect to these data sources right to bring that information back quickly so they don't have to replicate that information across different systems or different applications right and then to be able to provide that those answers back quickly also allowing for seamless access to from the analysts that are running running full speed right try and find the answers as quickly as they find great okay and I want to get into that sort of how news Steve let me go to you one of the things that we talked about earlier was just infusing this this mindset of a data cult and thinking about data as a service so talk a little bit about how you got started what was the starting NICUs through that sure I think the biggest thing for us there is to change that mindset from data being just for reporting or things that have happened in the past to do some insights on us and some data that already existed well we've tried to shift the mentality there is to start to use data and use that into our actual applications so that we're providing those insight in real time through the applications as they're consumed helping with customer experience helping with our personalization and an optimization of our application the way we've started down that path or kind of the journey that we're still on was to get the foundation laid birch so part of that has been making sure we have access to all that data whether it's through virtualization like vic talked about or whether it's through having more of the the data selected in a data like that that where we have all of that foundational data available as opposed to waiting for people to ask for it that's been the biggest culture shift for us is having that availability of data to be ready to be able to provide those insights as opposed to having to make the businesses or the application or asked for that day Oh Kailyn when I first met into pulp andari the idea wobble he paid up there yeah I was asking him okay where does a what's the role of that at CBO and and he mentioned a number of things but two of the things that stood out is you got to understand how data affect the monetization of your company that doesn't mean you know selling the data what role does it play and help cut cost or ink revenue or productivity or no customer service etc the other thing he said was you've got a align with the lines of piss a little sounded good and this is several years ago and IBM took it upon itself Greek its own champagne I was gonna say you know dogfooding whatever but it's not easy just flip a switch and an infuse a I and automate the data pipeline you guys had to go you know some real of pain to get there and you did you were early on you took some arrows and now you're helping your customers better on thin debt but talk about some of the use cases that where you guys have applied this obviously the biggest organization you know one of the biggest in the world the real challenge is they're sure I'm happy today you know we've been on this journey for about four years now so we stood up our first book to get office 2016 and you're right it was all about getting what data strategy offered and executed internally and we want to be very transparent because as you've mentioned you know a lot of challenges possible think differently about the value and so as we wrote that data strategy at that time about coming to enterprise and then we quickly of pivoted to see the real opportunity and value of infusing AI across all of our needs were close to your question on a couple of specific use cases I'd say you know we invested that time getting that platform built and implemented and then we were able to take advantage of that one particular example that I've been really excited about I have a practitioner on my team who's a supply chain expert and a couple of years ago he started building out supply chain solution so that we can better mitigate our risk in the event of a natural disaster like the earthquake hurricane anywhere around the world and be cuz we invest at the time and getting the date of pipelines right getting that all of that were created and cleaned and the quality of it we were able to recently in recent weeks add the really critical Kovach 19 data and deliver that out to our employees internally for their preparation purposes make that available to our nonprofit partners and now we're starting to see our first customers take advantage too with the health and well-being of their employees mine so that's you know an example I think where and I'm seeing a lot of you know my clients I work with they invest in the data and AI readiness and then they're able to take advantage of all of that work work very quickly in an agile fashion just spin up those out well I think one of the keys there who Kaelin is that you know we can talk about that in a covet 19 contact but it's that's gonna carry through that that notion of of business resiliency is it's gonna live on you know in this post pivot world isn't it absolutely I think for all of us the importance of investing in the business continuity and resiliency type work so that we know what to do in the event of either natural disaster or something beyond you know it'll be grounded in that and I think it'll only become more important for us to be able to act quickly and so the investment in those platforms and approach that we're taking and you know I see many of us taking will really be grounded in that resiliency so Vic and Steve I want to dig into this a little bit because you know we use this concept of data op we're stealing from DevOps and there are similarities but there are also differences now let's talk about the data pipeline if you think about the data pipeline as a sort of quasi linear process where you're investing data and you might be using you know tools but whether it's Kafka or you know we have a favorite who will you have and then you're transforming that that data and then you got a you know discovery you got to do some some exploration you got to figure out your metadata catalog and then you're trying to analyze that data to get some insights and then you ultimately you want to operationalize it so you know and and you could come up with your own data pipeline but generally that sort of concept is is I think well accepted there's different roles and unlike DevOps where it might be the same developer who's actually implementing security policies picking it the operations in in data ops there might be different roles and fact very often are there's data science there's may be an IT role there's data engineering there's analysts etc so Vic I wonder if you could you could talk about the challenges in in managing and automating that data pipeline applying data ops and how practitioners can overcome them yeah I would say a perfect example would be a client that I was just recently working for where we actually took a team and we built up a team using agile methodologies that framework right we're rapidly ingesting data and then proving out data's fit for purpose right so often now we talk a lot about big data and that is really where a lot of industries are going they're trying to add an enrichment to their own data sources so what they're doing is they're purchasing these third-party data sets so in doing so right you make that initial purchase but what many companies are doing today is they have no real way to vet that so they'll purchase the information they aren't going to vet it upfront they're going to bring it into an environment there it's going to take them time to understand if the data is of quality or not and by the time they do typically the sales gone and done and they're not going to ask for anything back but we were able to do it the most recent claim was use an instructure data source right bring that and ingest that with modelers using this agile team right and within two weeks we were able to bring the data in from the third-party vendor what we considered rapid prototyping right be able to profile the data understand if the data is of quality or not and then quickly figure out that you know what the data's not so in doing that we were able to then contact the vendor back tell them you know it sorry the data set up to snuff we'd like our money back we're not gonna go forward with it that's enabling businesses to be smarter with what they're doing with 30 new purchases today as many businesses right now um as much as they want to rely on their own data right they actually want to rely on cross the data from third-party sources and that's really what data Ops is allowing us to do it's allowing us to think at a broader a higher level right what to bring the information what structures can we store them in that they don't necessarily have to be modeled because a modeler is great right but if we have to take time to model all the information before we even know we want to use it that's gonna slow the process now and that's slowing the business down the business is looking for us to speed up all of our processes a lot of what we heard in the past raised that IP tends to slow us down and that's where we're trying to change that perception in the industry is no we're actually here to speed you up we have all the tools and technologies to do so and they're only getting better I would say also on data scientists right that's another piece of the pie for us if we can bring the information in and we can quickly catalog it in a metadata and burn it bring in the information in the backend data data assets right and then supply that information back to scientists gone are the days where scientists are going and asking for connections to all these different data sources waiting days for access requests to be approved just to find out that once they figure out how it with them the relationship diagram right the design looks like in that back-end database how to get to it write the code to get to it and then figure out this is not the information I need that Sally next to me right fold me the wrong information that's where the catalog comes in that's where due to absent data governance having that catalog that metadata management platform available to you they can go into a catalog without having to request access to anything quickly and within five minutes they can see the structures what if the tables look like what did the fields look like are these are these the metrics I need to bring back answers to the business that's data apps it's allowing us to speed up all of that information you know taking stuff that took months now down two weeks down two days down two hours so Steve I wonder if you could pick up on that and just help us understand what data means you we talked about earlier in our previous conversation I mentioned it upfront is this notion of you know the demand for for data access is it was through the roof and and you've gone from that to sort of more of a self-service environment where it's not IT owning the data it's really the businesses owning the data but what what is what is all this data op stuff meaning in your world sure I think it's very similar it's it's how do we enable and get access to that clicker showing the right controls showing the right processes and and building that scalability and agility and into all of it so that we're we're doing this at scale it's much more rapidly available we can discover new data separately determine if it's right or or more importantly if it's wrong similar to what what Vic described it's it's how do we enable the business to make those right decisions on whether or not they're going down the right path whether they're not the catalog is a big part of that we've also introduced a lot of frameworks around scale so just the ability to rapidly ingest data and make that available has been a key for us we've also focused on a prototyping environment so that sandbox mentality of how do we rapidly stand those up for users and and still provide some controls but have provide that ability for people to do that that exploration what we're finding is that by providing the platform and and the foundational layers that were we're getting the use cases to sort of evolve and come out of that as opposed to having the use cases prior to then go build things from we're shifting the mentality within the organization to say we don't know what we need yet let's let's start to explore that's kind of that data scientist mentality and culture it more of a way of thinking as opposed to you know an actual project or implement well I think that that cultural aspect is important of course Caitlin you guys are an AI company or at least that you know part of what you do but you know you've you for four decades maybe centuries you've been organized around different things by factoring plant but sales channel or whatever it is but-but-but-but how has the chief data officer organization within IBM been able to transform itself and and really infuse a data culture across the entire company one of the approaches you know we've taken and we talk about sort of the blueprint to drive AI transformation so that we can achieve and deliver these really high value use cases we talked about the data the technology which we've just pressed on with organizational piece of it duration are so important the change management enabling and equipping our data stewards I'll give one a civic example that I've been really excited about when we were building our platform and starting to pull districting structured unstructured pull it in our ADA stewards are spending a lot of time manually tagging and creating business metadata about that data and we identified that that was a real pain point costing us a lot of money valuable resources so we started to automate the metadata and doing that in partnership with our deep learning practitioners and some of the models that they were able to build that capability we pushed out into our contacts our product last year and one of the really exciting things for me to see is our data stewards who be so value exporters and the skills that they bring have reported that you know it's really changed the way they're able to work it's really sped up their process it's enabled them to then move on to higher value to abilities and and business benefits so they're very happy from an organizational you know completion point of view so I think there's ways to identify those use cases particularly for taste you know we drove some significant productivity savings we also really empowered and hold our data stewards we really value to make their job you know easier more efficient and and help them move on to things that they are more you know excited about doing so I think that's that you know another example of approaching taken yes so the cultural piece the people piece is key we talked a little bit about the process I want to get into a little bit into the tech Steve I wonder if you could tell us you know what's it what's the tech we have this bevy of tools I mentioned a number of them upfront you've got different data stores you've got open source pooling you've got IBM tooling what are the critical components of the technology that people should be thinking about tapping in architecture from ingestion perspective we're trying to do a lot of and a Python framework and scaleable ingestion pipe frameworks on the catalog side I think what we've done is gone with IBM PAC which provides a platform for a lot of these tools to stay integrated together so things from the discovery of data sources the cataloging the documentation of those data sources and then all the way through the actual advanced analytics and Python models and our our models and the open source ID combined with the ability to do some data prep and refinery work having that all in an integrated platform was a key to us for us that the rollout and of more of these tools in bulk as opposed to having the point solutions so that's been a big focus area for us and then on the analytic side and the web versus IDE there's a lot of different components you can go into whether it's meal soft whether it's AWS and some of the native functionalities out there you mentioned before Kafka and Anissa streams and different streaming technologies those are all the ones that are kind of in our Ketil box that we're starting to look at so and one of the keys here is we're trying to make decisions in as close to real time as possible as opposed to the business having to wait you know weeks or months and then by the time they get insights it's late and really rearview mirror so Vic your focus you know in your career has been a lot on data data quality governance master data management data from a data quality standpoint as well what are some of the key tools that you're familiar with that you've used that really have enabled you operationalize that data pipeline you know I would say I'm definitely the IBM tools I have the most experience with that also informatica though as well those are to me the two top players IBM definitely has come to the table with a suite right like Steve said cloud pack for data is really a one-stop shop so that's allowing that quick seamless access for business user versus them having to go into some of the previous versions that IBM had rolled out where you're going into different user interfaces right to find your information and that can become clunky it can add the process it can also create almost like a bad taste and if in most people's mouths because they don't want to navigate from system to system to system just to get their information so cloud pack to me definitely brings everything to the table in one in a one-stop shop type of environment in for me also though is working on the same thing and I would tell you that they haven't come up with a solution that really comes close to what IBM is done with cloud pack for data I'd be interested to see if they can bring that on the horizon but really IBM suite of tools allows for profiling follow the analytics write metadata management access to db2 warehouse on cloud those are the tools that I've worked in my past to implement as well as cloud object store to bring all that together to provide that one stop that at Northwestern right we're working right now with belieber I think calibra is a great set it pool are great garments catalog right but that's really what it's truly made for is it's a governance catalog you have to bring some other pieces to the table in order for it to serve up all the cloud pack does today which is the advanced profiling the data virtualization that cloud pack enables today the machine learning at the level where you can actually work with our and Python code and you put our notebooks inside of pack that's some of this the pieces right that are missing in some of the under vent other vendor schools today so one of the things that you're hearing here is the theme of openness others addition we've talked about a lot of tools and not IBM tools all IBM tools there there are many but but people want to use what they want to use so Kaitlin from an IBM perspective what's your commitment the openness number one but also to you know we talked a lot about cloud packs but to simplify the experience for your client well and I thank Stephen Victoria for you know speaking to their experience I really appreciate feedback and part of our approach has been to really take one the challenges that we've had I mentioned some of the capabilities that we brought forward in our cloud platform data product one being you know automating metadata generation and that was something we had to solve for our own data challenges in need so we will continue to source you know our use cases from and grounded from a practitioner perspective of what we're trying to do and solve and build and the approach we've really been taking is co-creation line and that we roll these capability about the product and work with our customers like Stephen light victorious you really solicit feedback to product route our dev teams push that out and just be very open and transparent I mean we want to deliver a seamless experience we want to do it in partnership and continue to solicit feedback and improve and roll out so no I think that will that has been our approach will continue to be and really appreciate the partnerships that we've been able to foster so we don't have a ton of time but I want to go to practitioners on the panel and ask you about key key performance indicators when I think about DevOps one of the things that we're measuring is the elapsed time the deploy applications start finished where we're measuring the amount of rework that has to be done the the quality of the deliverable what are the KPIs Victoria that are indicators of success in operationalizing date the data pipeline well I would definitely say your ability to deliver quickly right so how fast can you deliver is that is that quicker than what you've been able to do in the past right what is the user experience like right so have you been able to measure what what the amount of time was right that users are spending to bring information to the table in the past versus have you been able to reduce that time to delivery right of information business answers to business questions those are the key performance indicators to me that tell you that the suite that we've put in place today right it's providing information quickly I can get my business answers quickly but quicker than I could before and the information is accurate so being able to measure is it quality that I've been giving that I've given back or is this not is it the wrong information and yet I've got to go back to the table and find where I need to gather that from from somewhere else that to me tells us okay you know what the tools we've put in place today my teams are working quicker they're answering the questions they need to accurately that is when we know we're on the right path Steve anything you add to that I think she covered a lot of the people components the around the data quality scoring right for all the different data attributes coming up with a metric around how to measure that and and then showing that trend over time to show that it's getting better the other one that we're doing is just around overall date availability how how much data are we providing to our users and and showing that trend so when I first started you know we had somewhere in the neighborhood of 500 files that had been brought into the warehouse and and had been published and available in the neighborhood of a couple thousand fields we've grown that into weave we have thousands of cables now available so it's it's been you know hundreds of percent in scale as far as just the availability of that data how much is out there how much is is ready and available for for people to just dig in and put into their their analytics and their models and get those back into the other application so that's another key metric that we're starting to track as well so last question so I said at the top that every application is gonna need to be infused with AI this decade otherwise that application not going to be as competitive as it could be and so for those that are maybe stuck in their journey don't really know where to get started I'll start with with Caitlin and go to Victoria and then and then even bring us home what advice would you give the people that need to get going on this my advice is I think you pull the folks that are either producing or accessing your data and figure out what the rate is between I mentioned some of the data management challenges we were seeing this these processes were taking weeks and prone to error highly manual so part was ripe for AI project so identifying those use cases I think that are really causing you know the most free work and and manual effort you can move really quickly and as you build this platform out you're able to spin those up on an accelerated fashion I think identifying that and figuring out the business impact are able to drive very early on you can get going and start really seeing the value great yeah I would actually say kids I hit it on the head but I would probably add to that right is the first and foremost in my opinion right the importance around this is data governance you need to implement a data governance at an enterprise level many organizations will do it but they'll have silos of governance you really need an interface I did a government's platform that consists of a true framework of an operational model model charters right you have data domain owners data domain stewards data custodians all that needs to be defined and while that may take some work in in the beginning right the payoff down the line is that much more it's it it's allowing your business to truly own the data once they own the data and they take part in classifying the data assets for technologists and for analysts right you can start to eliminate some of the technical debt that most organizations have acquired today they can start to look at what are some of the systems that we can turn off what are some of the systems that we see valium truly build out a capability matrix we can start mapping systems right to capabilities and start to say where do we have wares or redundancy right what can we get rid of that's the first piece of it and then the second piece of it is really leveraging the tools that are out there today the IBM tools some of the other tools out there as well that enable some of the newer next-generation capabilities like unit nai right for example allowing automation for automation which right for all of us means that a lot of the analysts that are in place today they can access the information quicker they can deliver the information accurately like we've been talking about because it's been classified that pre works being done it's never too late to start but once you start that it just really acts as a domino effect to everything else where you start to see everything else fall into place all right thank you and Steve bring us on but advice for your your peers that want to get started sure I think the key for me too is like like those guys have talked about I think all everything they said is valid and accurate thing I would add is is from a starting perspective if you haven't started start right don't don't try to overthink that over plan it it started just do something and and and start the show that progress and value the use cases will come even if you think you're not there yet it's amazing once you have the national components there how some of these things start to come out of the woodwork so so it started it going may have it have that iterative approach to this and an open mindset it's encourage exploration and enablement look your organization in the eye to say why are their silos why do these things like this what are our problem what are the things getting in our way and and focus and tackle those those areas as opposed to trying to put up more rails and more boundaries and kind of encourage that silo mentality really really look at how do you how do you focus on that enablement and then the last comment would just be on scale everything should be focused on scale what you think is a one-time process today you're gonna do it again we've all been there you're gonna do it a thousand times again so prepare for that prepare forever that you're gonna do everything a thousand times and and start to instill that culture within your organization a great advice guys data bringing machine intelligence an AI to really drive insights and scaling with a cloud operating model no matter where that data live it's really great to have have three such knowledgeable practitioners Caitlyn Toria and Steve thanks so much for coming on the cube and helping support this panel all right and thank you for watching everybody now remember this panel was part of the raw material that went into a crowd chat that we hosted on May 27th Crouch at net slash data ops so go check that out this is Dave Volante for the cube thanks for watching [Music]

from the cube studios in Palo Alto in Boston connecting with thought leaders all around the world this is a cube conversation hi Bri welcome back this is Dave Volante and welcome to this special presentation made possible by IBM we're talking about data op data ops in Acton Steve Lucas here he's the senior vice president and director of data management at Associated Bank be great to see how are things going and in Wisconsin all safe we're doing well we're staying safe staying healthy thanks for having me Dave yeah you're very welcome so Associated Bank and regional bank Midwest to cover a lot of the territories not just Wisconsin but another number of other states around there retail commercial lending real estate offices stuff I think the largest bank in in Wisconsin but tell us a little bit about your business in your specific role sure yeah no it's a good intro we're definitely largest bank at Corvis concen and then we have branches in the in the Upper Midwest area so Minnesota Illinois Wisconsin our primary locations my role at associated I'm director data management so been with the bank a couple of years now and really just focused on defining our data strategy as an overall everything from data ingestion through consumption of data and analytics all the way through and then I'm also the data governance components and keeping the controls and the rails in place around all of our data in its usage so financial services obviously one of the more cutting-edge industries in terms of their use of technology not only are you good negotiators but you you often are early adopters you guys were on the Big Data bandwagon early a lot of financial services firms we're kind of early on in Hadoop but I wonder if you could tell us a little bit about sort of the business drivers and and where's the poor the pressure point that are informing your digital strategy your your data and data op strategy sure yeah I think that one of the key areas for us is that we're trying to shift from more of a reactive mode into more of a predictive prescriptive mode from a data and analytics perspective and using our data to infuse and drive more business decisions but also to infuse it in actual applications and customer experience etc so we have a wealth of data at our fingertips we're really focused on starting to build out that data link style strategy make sure that we're kind of ahead of the curve as far as trying to predict what our end users are going to need and some of the advanced use cases we're going to have before we even know that they actually exist right so it's really trying to prepare us for the future and what's next and and then abling and empowering the business to be able to pivot when we need to without having everything perfect that they prescribed and and ready for what if we could talk about a little bit about the data journey I know it's kind of a buzzword but in my career as a independent observer and analyst I've kind of watched the promise of whether it was decision support systems or enterprise data warehouse you know give that 360 degree view of the business the the real-time nature the the customer intimacy all that in and up until sort of the recent digital you know meme I feel as though the industry hasn't lived up to that promise so I wonder if you could take us through the journey and tell us sort of where you came from and where you are today and I really want to sort of understand some of the successes they've had sure no that's a that's a great point nice I feel like as an industry I think we're at a point now where the the people process technology have sort of all caught up to each other right I feel that that real-time streaming analytics the data service mentality just leveraging web services and API is more throughout our organization in our industry as a whole I feel like that's really starting to take shape right now and and all the pieces of that puzzle have come together so kind of where we started from a journey perspective it was it was very much if your your legacy reporting data warehouse mindset of tell me tell me the data elements that you think you're going to need we'll figure out how do we map those in and form them we'll figure out how to get those prepared for you and that whole lifecycle that waterfall mentality of how do we get this through the funnel and get it to users quality was usually there the the enablement was still there but it was missing that that rapid turnaround it was also missing the the what's next right than what you haven't thought of and almost to a point of just discouraging people from asking for too many things because it got too expensive it got too hard to maintain there was some difficulty in that space so some of the things that we're trying to do now is build that that enablement mentality of encouraging people to ask for everything so when we bring out new systems - the bank is no longer an option as far as how much data they're going to send to us right we're getting all of the data we're going to we're going to bring that all together for people and then really starting to figure out how can this data now be used and and we almost have to push that out and infuse it within our organization as opposed to waiting for it to be asked for so I think that all of the the concepts so that bringing that people process and then now the tools and capabilities together has really started to make a move for us and in the industry I mean it's really not an uncommon story right you had a traditional data warehouse system you had you know some experts that you had to go through to get the data the business kind of felt like it didn't own the data you know it felt like it was imposing every time it made a request or maybe it was frustrated because it took so long and then by the time they got the data perhaps you know the market had shifted so it create a lot of frustration and then to your point but but it became very useful as a reporting tool and that was kind of this the sweet spot so so how did you overcome that and you know get to where you are today and you know kind of where are you today I was gonna say I think we're still overcoming that we'll see it'll see how this all goes right I think there's there's a couple of things that you know we've started to enable first off is just having that a concept of scale and enablement mentality and everything that we do so when we bring systems on we bring on everything we're starting to have those those components and pieces in place and we're starting to build more framework base reusable processes and procedures so that every ask is not brand new it's not this reinvent the wheel and resolve for for all that work so I think that's helped if expedite our time to market and really get some of the buy-in and support from around the organization and it's really just finding the right use cases and finding the different business partners to work with and partner with so that you help them through their journey as well is there I'm there on a similar roadmap and journey for for their own life cycles as well in their product element or whatever business line there so from a process standpoint that you kind of have to jettison the you mentioned waterfall before and move to a more being an agile approach did it require different different skill sets talk about the process and the people side of yeah it's been a it's been a shift we've tried to shift more towards I wouldn't call us more formal agile I would say we're a little bit more lean from a an iterative backlog type of approach right so what are you putting that work together in queues and having the queue of B reprioritized working with the business owners to help through those things has been a key success criteria for us and how we start to manage that work as opposed to opening formal project requests and and having all that work have to funnel through some of the old channels that like you mentioned earlier kind of distracted a little bit from from the way things had been done in the past and added some layers that people felt potentially wouldn't be necessary if they thought it was a small ask in their eyes you know I think it also led to a lot of some of the data silos and and components that we have in place today in the industry and I don't think our company is alone and having data silos and components of data in different locations but those are there for a reason though those were there because they're they're filling a need that has been missing or a gap in the solution so what we're trying to do is really take that to heart and evaluate what can we do to enable those mindsets and those mentalities and find out what was the gap and why did they have to go get a siloed solution or work around operations and technology and the channels that had been in place what would you say well your biggest challenges in getting from point A to point B point B being where you are today there were challenges on each of the components of the pillar right so people process technology people are hard to change right men behavioral type changes has been difficult that there's components of that that definitely has been in place same with the process side right so so changing it into that backlog style mentality and working with the users and having more that be sort of that maintenance type support work is is a different call culture for our organization and traditional project management and then the tool sets right the the tools and capabilities we had to look in and evaluate what tools do we need to Mabel this behavior in this mentality how do we enable more self-service the exploration how do we get people the data that they need when they need it and empower them to use so maybe you could share with us some of the outcomes and I know it's yeah we're never done in this business but but thinking about you know the investments that you've made in intact people in reprocessing you know the time it takes to get leadership involved what has been so far anyway the business outcome and you share any any metrics or it is sort of subjective a guidance I yeah I think from a subjective perspective the some of the biggest things for us has just been our ability to to truly start to have that very 60 degree view of the customer which we're probably never going to get they're officially right there's there everyone's striving for that but the ability to have you know all of that data available kind of at our fingertips and have that all consolidated now into one one location one platform and start to be that hub that starts to redistribute that data to our applications and infusing that out has been a key component for us I think some of the other big kind of components are differentiators for us and value that we can show from an organizational perspective we're in an M&A mode right so we're always looking from a merger and acquisition perspective our the model that we've built out from a data strategy perspective has proven itself useful over and over now in that M&A mentality of how do you rapidly ingest new data sets it had understood get it distributed to the right consumers it's fit our model exactly and and it hasn't been an exception it's been just part of our overall framework for how we get that data and it wasn't anything new that we had to do different because it was M&A just timelines were probably a little bit more expedited the other thing that's been interesting in some of the world that were in now right from a a Kovach perspective and having a pivot and start to change some of the way we do business and some of the PPP loans and and our business models sort of had to change overnight and our ability to work with our different lines of business and get them the data they need to help drive those decisions was another scenario where had we not had the foundational components there in the platform there to do some of this if we would have spun a little bit longer so your data ops approach I'm gonna use that term helped you in this in this kovat situation I mean you had the PPE you had you know of slew of businesses looking to get access to that money you had uncertainty with regard to kind of what the rules of the game were what you was the bank you had a Judah cape but you it was really kind of opaque in terms of what you had to do the volume of loans had to go through the roof in the time frame it was like within days or weeks that you had to provide these so I wonder if we could talk about that a little bit and how you're sort of approach the data helped you be prepared for that yeah no it was a race I mean the bottom line was it felt like a race right from from industry perspective as far as how how could we get this out there soon enough fast enough provide the most value to our customers our applications teams did a phenomenal job on enabling the applications to help streamline some of the application process for the loans themselves but from a data and reporting perspective behind the scenes we were there and we had some tools and capabilities and readiness to say we have the data now in our in our lake we can start to do some business driven decisions around all all of the different components of what's being processed on a daily basis from an application perspective versus what's been funded and how do those start to funnel all the way through doing some data quality checks and operational reporting checks to make sure that that data move properly and got booked in in the proper ways because of the rapid nature of how that was was all being done other covent type use cases as well we had some some different scenarios around different feed reporting and and other capabilities that the business wasn't necessarily prepared for we wouldn't have planned to have some of these types of things and reporting in place that we were able to give it because we had access to all the data because of these frameworks that we had put into place that we could pretty rapidly start to turn around some of those data some of those data points and analytics for us to make some some better decisions so given the propensity in the pace of M&A there has to be a challenge fundamentally in just in terms of data quality consistency governance give us the before and after you know before kind of before being the before the data ops mindset and after being kind of where you are today I think that's still a journey we're always trying to get better on that as well but the data ops mindset for us really has has shifted us to start to think about automation right pipelines that enablement a constant improvement and and how do we deploy faster deploy more consistently and and have the right capabilities in place when we need it so you know where some of that has come into place from an M&A perspective is it's really been around the building scale into everything that we do dezq real-time nature this scalability the rapid deployment models that we have in place is really where that starts to join forces and really become become powerful having having the ability to rapidly ingesting new data sources whether we know about it or not and then exposing that and having the tools and platforms be able to expose that to our users and enable our business lines whether it's covent whether it's M&A the use cases keep coming up right they we keep running into the same same concept which is how rapidly get people the data they need when they need it but still provide the rails and controls and make sure that it's governed and controllable on the way as well [Music] about the tech though wonder if we could spend some time on that I mean can you paint a picture of us so I thought what what what we're looking at here you've got you know some traditional IDI w's involved I'm sure you've got lots of data sources you you may be one of the zookeepers from the the Hadoop days with a lot of you know experimentation there may be some machine intelligence and they are painting a pic before us but sure no so we're kind of evolving some of the tool sets and capabilities as well we have some some generic kind of custom in-house build ingestion frameworks that we've started to build out for how to rapidly ingest and kind of script out the nature of of how we bring those data sources into play what we're what we've now started as well as is a journey down IBM compact product which is really gonna it's providing us that ability to govern and control all of our data sources and then start to enable some of that real-time ad hoc analytics and data preparation data shaping so some of the components that we're doing in there is just around that data discovery pointing that data sources rapidly running data profiles exposing that data to our users obviously very handy in the emanating space and and anytime you get new data sources in but then the concept of publishing that and leveraging some of the AI capabilities of assigning business terms in the data glossary and those components is another key component for us on the on the consumption side of the house for for data we have a couple of tools in place where Cognos shop we do a tableau from a data visualization perspective as well that what that were we're leveraging but that's where cloud pack is now starting to come into play as well from a data refinement perspective and giving the ability for users to actually go start to shape and prep their data sets all within that governed concept and then we've actually now started down the enablement path from an AI perspective with Python and R and we're using compact to be our orchestration tool to keep all that governed and controlled as well enable some some new AI models and some new technologies in that space we're actually starting to convert all of our custom-built frameworks into python now as well so we start to have some of that embedded within cloud pack and we can start to use some of the rails of those frameworks with it within them okay so you've got the ingest and ingestion side you've done a lot of automation it sounds like called the data profiling that's maybe what classification and automating that piece and then you've got the data quality piece the governance you got visualization with with tableau and and this kind of all fits together in a in an open quote unquote open framework is that right yeah I exactly I mean the the framework itself from our perspective where we're trying to keep the tools as as consistent as we can we really want to enable our users to have the tools that they need in the toolbox and and keep all that open what we're trying to focus on is making sure that they get the same data the same experience through whatever tool and mechanism that they're consuming from so that's where that platform mentality comes into place having compact in the middle to help govern all that and and reprovision some of those data sources out for us has it has been a key component for us well see if it sounds like you're you know making a lot of progress or you know so the days of the data temple or the high priest of data or the sort of keepers of that data really to more of a data culture where the businesses kind of feel ownership for their own data you believe self-service I think you've got confidence much more confident than the in the compliance and governance piece but bring us home just in terms of that notion of data culture and where you are and where you're headed no definitely I think that's that's been a key for us too as as part of our strategy is really helping we put in a strategy that helps define and dictate some of those structures and ownership and make that more clear some of the of the failures of the past if you will from an overall my monster data warehouse was around nobody ever owned it there was there wasn't you always ran that that risk of either the loudest consumer actually owned it or no one actually owned it what we've started to do with this is that Lake mentality and and having all that data ingested into our our frameworks the data owners are clear-cut it's who sends that data in what is the book record system for that source data we don't want a ability we don't touch it we don't transform it as we load it it sits there and available you own it we're doing the same mentality on the consumer side so we have we have a series of structures from a consumption perspective that all of our users are consuming our data if it's represented exactly how they want to consume it so again that ownership we're trying to take out a lot of that gray area and I'm enabling them to say yeah I own this I understand what I'm what I'm going after and and I can put the the ownership and the rule and rules and the stewardship around that as opposed to having that gray model in the middle that that that we never we never get but I guess to kind of close it out really the the concept for us is enabling people and end-users right giving them the data that they need when they need it and it's it's really about providing the framework and then the rails around around doing that and it's not about building out a formal bill warehouse model or a formal lessor like you mentioned before some of the you know the ivory tower type concepts right it's really about purpose-built data sets getting the giving our users empowered with the data they need when they need it all the way through and fusing that into our applications so that the applications and provide the best user experiences and and use the data to our advantage all about enabling the business I got a shove all I have you how's that IBM doing you know as a as a partner what do you like what could they be doing better to make your life easier sure I think I think they've been a great partner for us as far as that that enablement mentality the cloud pack platform has been a key for us we wouldn't be where we are without that tool said I our journey originally when we started looking at tools and modernization of our staff was around data quality data governance type components and tools we now because of the platform have released our first Python I models into the environment we have our studio capabilities natively because of the way that that's all container is now within cloud back so we've been able to enable new use cases and really advance us where we would have a time or a lot a lot more technologies and capabilities and then integrate those ourselves so the ability to have that all done has or and be able to leverage that platform has been a key to helping us get some of these roles out of this as quickly as we have as far as a partnership perspective they've been great as far as listening to what what the next steps are for us where we're headed what can we what do we need more of what can they do to help us get there so it's it's really been an encouraging encouraging environment I think they as far as what can they do better I think it's just keep keep delivering write it delivery is ping so keep keep releasing the new functionality and features and keeping the quality of the product intact well see it was great having you on the cube we always love to get the practitioner angle sounds like you've made a lot of progress and as I said when we're never finished in this industry so best of luck to you stay safe then and thanks so much for for sharing appreciate it thank you all right and thank you for watching everybody this is Dave Volante for the cube data ops in action we got the crowd chat a little bit later get right there but right back right of this short break [Music] [Music]

from the cube studios in Palo Alto in Boston connecting with thought leaders all around the world this is a cube conversation hi everybody this is David on tape with the cube and welcome to the special digital presentation we're really digging into how IBM is operationalizing and automating the AI and data pipeline not only for its clients but also for itself and with me is Julie Lochner who looks after offering management and IBM's data and AI portfolio Julie great to see you again okay great to be here thank you talk a little bit about the role you have here at IBM sure so my responsibility in offering management in the data and AI organization is really twofold one is I lead a team that implements all of the back-end processes really the operations behind anytime we deliver a product from the data AI team to the market so think about all of the release cycle management pricing product management discipline etc the other roles that I play is really making sure that um we are working with our customers and making sure they have the best customer experience and a big part of that is developing the data ops methodology it's something that I needed internally from my own line of business execution but it's now something that our customers are looking for to implement in their shops as well well good I really want to get into that and so let's let's start with data ops I mean I think you know a lot of people are familiar with DevOps not maybe not everybody's familiar with the data Ops what do we need to know about data well I mean you bring up the point that everyone knows DevOps and and then in fact I think you know what data Ops really does is bring a lot of the benefits that DevOps did for application development to the data management organizations so when we look at what is data ops it's a data management it's a it's a data management set of principles that helps organizations bring business ready data to their consumers quickly it takes it borrows from DevOps similarly where you have a data pipeline that associates a business value requirement I have this business initiative it's gonna drive this much revenue or this much cost savings this is the data that I need to be able to deliver it how do I develop that pipeline and map to the data sources know what data it is know that I can trust it so ensuring that it has the right quality that I'm actually using the data that it was meant for and then put it to use so in in history most dated management practices deployed a waterfall like methodology or implementation methodology and what that meant is all the data pipeline projects were implemented serially and it was dawn based on potentially a first-in first-out program management office with a DevOps mental model and the idea of being able to slice through all of the different silos that's required to collect the data to organize it to integrate it to validate its quality to create those data integration pipelines and then present it to the dashboard like if it's a Cognos dashboard for a operational process or even a data science team that whole end-to-end process gets streamlined through what we're calling data ops methodology so I mean as you well know we've been following this market since the early days of a dupe and people struggle with their data pipelines it's complicated for them there's a raft of tools and and and they spend most of their time wrangling data preparing data improving data quality different roles within the organization so it sounds like you know to borrow from from DevOps data OPS's is all about REME lining that data pipeline helping people really understand and communicate across end to end as you're saying but but what's the ultimate business outcome that you're trying to drive so when you think about projects that require data to again cut cost to automate a business process or drive new revenue initiatives how long does it take to get from having access to the data to making it available that duration for every time delay that is spent wasted trying to connect to data sources trying to find subject matter experts that understand what the data means and can verify its quality like all of those steps along those different teams and different disciplines introduces delay in delivering high quality data fast so the business value of data Ops is always associated with something that the business is trying to achieve but with a time element so if it's for every day we don't have this data to make a decision we're either making money or losing money that's the value proposition of data ops so it's about taking things that people are already doing today and figuring out the quickest way to do it through automation through workflows and just cutting through all of the political barriers that often happens when these data's cross different organizational boundaries yeah so speed time to insights is critical but to in and then you know with DevOps you're really bringing together the skill sets into sort of you know one super dev or one super ops it sounds with data ops it's really more about everybody understanding their role and having communication and line-of-sight across the entire organization it's not trying to make everybody a superhuman data person it's the whole it's the group it's the team effort really it's really a team game here isn't it well that's a big part of it so just like any type of practice there's people aspects process aspects and technology right so people process technology and while you're you're describing it like having that super team that knows everything about the data the only way that's possible is if you have a common foundation of metadata so we've seen a surgeons in the data catalog market and last you know six seven years and what what the what that the innovation in the data catalog market has actually enabled us to be able to drive more data ops pipelines meaning as you identify data assets you've captured the metadata you capture its meaning you capture information that can be shared whether they're stakeholders it really then becomes more of a essential repository for people to really quickly know what data they have really quickly understand what it means in its quality and very quickly with the right proper authority like privacy rules included put it to use for models you know dashboards operational processes okay and and we're gonna talk about some examples and one of them of course is ibm's own internal example but but help us understand where you advise clients to start I want to get into it where do I get started yeah I mean so traditionally what we've seen with these large data management data governance programs is that sometimes our customers feel like this is a big pill to swallow and what we've said is look there's an opera there's an opportunity here to quickly define a small project align it to a high-value business initiative target something that you can quickly gain access to the data map out these pipelines and create a squad of skills so it includes a person with DevOps type programming skills to automate an instrument a lot of the technology a subject matter expert who understands the data sources and its meaning a line of business executive who can translate bringing that information to the business project and associating with business value so when we say how do you get started we've developed a I would call it a pretty basic maturity model to help organizations figure out where are they in terms of the technology where are they in terms of organizationally knowing who the right people should be involved in these projects and then from a process perspective we've developed some pretty prescriptive project plans that help you nail down what are the data elements that are critical for this business business initiative and then we have for each role what their jobs are to consolidate the datasets map them together and present them to the consumer we find that six-week projects typically three sprints are perfect times to be able to in a timeline to create one of these very short quick win projects take that as an opportunity to figure out where your bottlenecks are in your own organization where your skill shortages are and then use the outcome of that six-week sprint to then focus on filling in gaps kick off the next project and iterate celebrate the success and promote the success because it's typically tied to a business value to help them create momentum for the next one all right that's awesome I want to now get into some examples I mean or you're we're both massachusetts-based normally you'd be in our studio and we'd be sitting here face-to-face obviously with kovat 19 in this crisis we're all sheltering in place you're up in somewhere in New England I happen to be in my studio believe it but I'm the only one here so relate this to kovat how would data ops or maybe you have a concrete example in in terms of how it's helped inform or actually anticipate and keep up-to-date with what's happening with building yeah well I mean we're all experiencing it I don't think there's a person on the planet who hasn't been impacted by what's been going on with this coded pandemic crisis so we started we started down this data obscurity a year ago I mean this isn't something that we just decided to implement a few weeks ago we've been working on developing the methodology getting our own organization in place so that we could respond the next time we needed to be able to you know act upon a data-driven decision so part of step one of our journey has really been working with our global chief data officer Interpol who I believe you have had an opportunity to meet with an interview so part of this year journey has been working with with our corporate organization I'm in the line of business organization where we've established the roles and responsibilities we've established the technology stack based on our cloud pack for data and Watson knowledge catalog so I use that as the context for now we're faced with a pandemic crisis and I'm being asked in my business unit to respond very quickly how can we prioritize the offerings that are gonna help those in critical need so that we can get those products out to market we can offer a you know 90-day free use for governments and Hospital agencies so in order for me to do that as a operations lead for our team I needed to be able to have access to our financial data I needed to have access to our product portfolio information I needed to understand our cloud capacity so in order for me to be able to respond with the offers that we recently announced you know you can take a look at some of the examples with our Watson citizen assistant program where I was able to provide the financial information required for us to make those products available for governments hospitals state agencies etc that's a that's a perfect example now to to set the stage back to the corporate global chief data office organization they implemented some technology that allowed us to ingest data automatically classify it automatically assign metadata automatically associate data quality so that when my team started using that data we knew what the status of that information was when we started to build our own predictive models and so that's a great example of how we've partnered with a corporate central organization and took advantage of the automated set of capabilities without having to invest in any additional resources or headcount and be able to release products within a matter of a couple of weeks and in that automation is a function of machine intelligence is that right and obviously some experience but but you couldn't you and I when we were consultants doing this by hand we couldn't have done this we could have done it at scale anyways it is it machine intelligence an AI that allows us to do this that's exactly right and as you know our organization is data and AI so we happen to have the a research and innovation teams that are building a lot of this technology so we have somewhat of an advantage there but you're right the alternative to what I've described is manual spreadsheets it's querying databases it's sending emails to subject matter experts asking them what this data means if they're out sick or on vacation you have to wait for them to come back and all of this was a manual process and in the last five years we've seen this data catalog market really become this augmented data catalog and that augmentation means it's automation through AI so with years of experience and natural language understanding we can comb through a lot of the metadata that's available electronically we can comb through unstructured data we can categorize it and if you have a set of business terms that have industry standard definitions through machine learning we can automate what you and I did as a consultant manually in a matter of seconds that's the impact the AI is had in our organization and now we're bringing this to the market and it's a it's a big part of where I'm investing my time both internally and externally is bringing these types of concepts and ideas to the market so I'm hearing first of all one of the things that strikes me is you've got multiple data sources and data lives everywhere you might have your supply chain data and your ERP maybe that sits on Prem you might have some sales data that's sitting in the SAS store in a cloud somewhere you might have you know a weather data that you want to bring in in theory anyway the more data that you have the better insights that you can gather assuming you've got the right data quality but so let me start with like where the data is right so so it sits anywhere you don't know where it's gonna be but you know you need it so that that's part of this right is being able to read it quickly yeah it's funny you bring it up that way I actually look a little differently it's when you start these projects the data was in one place and then by the time you get through the end of a project you find out that it's a cloud so the data location actually changes while we're in the middle of projects we have many or coming even during this this pandemic crisis we have many organizations that are using this as an opportunity to move to SAS so what was on Prem is now cloud but that shouldn't change the definition of the data it shouldn't change its meaning it might change how you connect to it um it might also change your security policies or privacy laws now all of a sudden you have to worry about where is that data physically located and am I allowed to share it across national boundaries right before we knew physically where it was so when you think about data ops data ops is a process that sits on top of where the data physically resides and because we're mapping metadata and we're looking at these data pipelines and automated workflows part of the design principles are to set it up so that it's independent of where it resides however you have to have placeholders in your metadata and in your tool chain where we oughta mating these workflows so that you can accommodate when the data decides to move because of corporate policy change from on-prem to cloud then that's a big part of what data Ops offers it's the same thing by the way for DevOps they've had to accommodate you know building in you know platforms as a service versus on from the development environments it's the same for data ops and you know the other part that strikes me and listening to you is scale and it's not just about you know scale with the cloud operating model it's also about what you're talking about is you know the auto classification the automated metadata you can't do that manually you've got to be able to do that in order to scale with automation that's another key part of data Ops is it not it's well it's a big part of the value proposition and a lot of a part of the business base right then you and I started in this business you know and Big Data became the thing people just move all sorts of data sets to these Hadoop clusters without capturing the metadata and so as a result you know in the last 10 years this information is out there but nobody knows what it means anymore so you can't go back with the army of people and have them query these data sets because a lot of the contact was lost but you can use automated technology you can use automated machine learning with natural under Snatcher Alang guaa Jing to do a lot of the heavy lifting for you and a big part of data ops workflows and building these pipelines is to do what we call management-by-exception so if your algorithms say you know 80% confident that this is a phone number and your organization has a you know low risk tolerance that probably will go to an exception but if you have a you know a match algorithm that comes back and says it's 99 percent sure this is an email address right and you I have a threshold that's 98% it will automate much of the work that we used to have to do manually so that's an example of how you can automate eliminate manual work and have some human interaction based on your risk threshold now that's awesome I mean you're right the no schema on right said I throw it into a data leg the data link becomes the data swap we all know that joke okay I want to understand a little bit and maybe you have some other examples of some of the use cases here but there's some of the maturity of where customers are I mean it seems like you got to start by just understanding what data you have cataloging it you're getting your metadata act in order but then you've got a you've got a data quality component before you can actually implement and get yet to insight so you know where our customers on the on the maturity model do you have any other examples that you can share yeah so when we look at our data ops maturity model we tried to simplify it I mentioned this earlier that we try to simplify it so that really anybody can get started they don't have to have a full governance framework implemented to take advantage of the benefits data ops delivers so what we did we said if you can categorize your data ops programs into really three things one is how well do you know your data do you even know what data you have the second one is and you trust it like can you trust its quality can you trust its meeting and the third one is can you put it to use so if you really think about it when you begin with what data do you know right the first step is you know how are you determining what data you know the first step is if you are using spreadsheets replace it with a data catalog if you have a department line of business catalog and you need to start sharing information with the departments then start expanding to an enterprise level data catalog now you mentioned data quality so the first step is do you even have a data quality program right have you even established what your criteria are for high quality data have you considered what your data quality score is comprised of have you mapped out what your critical data elements are to run your business most companies have done that for they're they're governed processes but for these new initiatives and when you identify I'm in my example with the Kovach crisis what products are we gonna help bring to market quickly I need to be able to find out what the critical data elements are and can I trust it have I even done a quality scan and have teams commented on its trustworthiness to be used in this case if you haven't done anything like that in your organization that might be the first place to start pick the critical data elements for this initiative assess its quality and then start to implement the workflows to remediate and then when you get to putting it to use there's several methods for making data available you know one is simply making a data Mart available to a small set of users that's what most people do well first they make a spreadsheet of the data available but then if they need to have multiple people access it that's when like a data Mart might make sense technology like data virtualization eliminates the need for you to move data as you're in this prototyping phase and that's a great way to get started it doesn't cost a lot of money to get a virtual query set up to see if this is the right join or the right combination of fields that are required for this use case eventually you'll get to the need to use a high performance ETL tool for data integration but Nirvana is when you really get to that self-service data prep where users can query a catalog and say these are the data sets I need it presents you a list of data assets that are available I can point and click at these columns I want as part of my you know data pipeline and I hit go and it automatically generates that output for data science use cases for a Cognos dashboard right that's the most mature model and being able to iterate on that so quickly that as soon as you get feedback that that data elements are wrong or you need to add something you can do it push button and that's where data observation to bring organizations to well Julie I think there's no question that this kovat crisis is accentuated the importance of digital you know we talk about digital transformation a lot and it's it's certainly real although I would say a lot of people that we talk to will say well you know not on my watch or I'll be retired before that all happens will this crisis is accelerating that transformation and data is at the heart of it you know digital means data and if you don't have your data you know story together and your act together then you're gonna you're not going to be able to compete and data ops really is a key aspect of that so you know give us a parting word all right I think this is a great opportunity for us to really assess how well we're leveraging data to make strategic decisions and if there hasn't been a more pressing time to do it it's when our entire engagement becomes virtual like this interview is virtual write everything now creates a digital footprint that we can leverage to understand where our customers are having problems where they're having successes you know let's use the data that's available and use data ops to make sure that we can iterate access that data know it trust it put it to use so that we can respond to those in need when they need it Julie Locker your incredible practitioner really hands-on really appreciate you coming on the Kuban and sharing your knowledge with us thank you okay thank you very much it was a pleasure to be here all right and thank you for watching everybody this is Dave Volante for the cube and we will see you next time [Music]

>>from the Cube Studios in >>Palo Alto and Boston connecting with thought leaders all around the world. This is a cube conversation. Everybody welcome this special digital presentation where we're covering the topic of data ops and specifically how IBM is really operationalize ing and automating the data pipeline with data office. And with me is Interpol Bhandari, who is the global chief data officer at IBM and Paul. It's always great to see you. Thanks for coming on. >>My pleasure. >>So, you know the standard throwaway question from guys like me And you know what keeps the chief data officer up at night? Well, I know what's keeping you up that night. It's coverted 19. How you >>doing? It's keeping keeping all of us. >>Yeah, for sure. Um, So how are you guys making out as a leader I'm interested in, You know, how you have responded would whether it's communications. Obviously you're doing much more stuff you remotely You're not on airplanes. Certainly like you used to be. But But what was your first move when you actually realized this was going to require a shift? >>Well, I think one of the first things that I did wants to test the ability of my organization, You work remotely. This was well before the the recommendations came in from the government just so that we wanted to be sure that this is something that we could pull off if there were extreme circumstances where even everybody was. And so that was one of the first things we did along with that. I think another major activity that's being boxed off is even that we have created this Central Data and AI platform for idea using our hybrid, multi cloud approach. How could that the adaptive very, very quickly help them look over the city? But those were the two big items that my team and my embarked on and again, like I said, this is before there was any recommendations from the government or even internally, within IBM. Have any recommendations be? We decided that we wanted to run ahead and make sure that we were ready to ready to operate in that fashion. And I believe a lot of my colleagues did the same. Yeah, >>there's a there's a conversation going on right now just around productivity hits that people may be taking because they really weren't prepared it sounds like you're pretty comfortable with the productivity impact that you're achieving. >>Oh, I'm totally comfortable with the politics. I mean, in fact, I will tell you that while we've gone down this spot, we've realized that in some cases the productivity is actually going to be better when people are working from home and they're able to focus a lot more on the work, you know, And this could. This one's the gamut from the nature of the jaw, where you know somebody who basically needs to be in the front of the computer and is remotely taking care of operations. You know, if they don't have to come in, their productivity is going to go up Somebody like myself who had a long drive into work, you know, which I would use a phone calls, but that that entire time it can be used a lot more productivity, locked in a lot more productive manner. So there is. We realized that there's going to be some aspect of productivity that will actually be helped by the situation. Why did you are able to deliver the services that you deliver with the same level of quality and satisfaction that you want Now there were certain other aspect where you know the whole activity is going to be effective. So you know my team. There's a lot off white boarding that gets done there lots off informal conversations that spot creativity. But those things are much harder to replicate in a remote and large. So we've got a sense off. You know where we have to do some work? Well, things together. This is where we're actually going to be mobile. But all in all, they're very comfortable that we can pull this off. >>That's great. I want to stay on Cove it for a moment and in the context of just data and data ops, and you know why Now, obviously, with a crisis like this, it increases the imperative to really have your data act together. But I want to ask you both specifically as it relates to covert, why Data office is so important. And then just generally, why at this this point in time, >>So, I mean, you know, the journey we've been on. Thank you. You know, when I joined our data strategy centered around cloud data and ai, mainly because IBM business strategy was around that, and because there wasn't the notion off AI and Enterprise, right, there was everybody understood what AI means for the consumer. But for the enterprise, people don't really understand. Well, what a man. So our data strategy became one off, actually making IBM itself into an AI and and then using that as a showcase for our clients and customers who look a lot like us, you make them into AI. And in a nutshell, what that translated to was that one had two in few ai into the workflow off the key business processes off enterprise. So if you think about that workflow is very demanding, right, you have to be able to deliver. They did not insights on time just when it's needed. Otherwise, you can essentially slow down the whole workflow off a major process within an end. But to be able to pull all that off you need to have your own data works very, very streamlined so that a lot of it is automated and you're able to deliver those insights as the people who are involved in the work floor needed. So we've spent a lot of time while we were making IBM into any I enterprise and infusing AI into our key business processes into essentially a data ops pipeline that was very, very streamlined, which then allowed us to do very quickly adapt do the over 19 situation and I'll give you one specific example that will go to you know how one would someone would essentially leverage that capability that I just talked about to do this. So one of the key business processes that we have taken a map, it was our supply chain. You know, if you're a global company and our supply chain is critical, you have lots of suppliers, and they are all over the globe. And we have different types of products so that, you know, has a multiplication factors for each of those, you have additional suppliers and you have events. You have other events, you have calamities, you have political events. So we have to be able to very quickly understand the risks associated with any of those events with regard to our supply chain and make appropriate adjustments on the fly. So that was one off the key applications that we built on our central data. And as Paul about data ops pipeline. That meant we ingest the ingestion off those several 100 sources of data not to be blazingly fast and also refresh very, very quickly. Also, we have to then aggregate data from the outside from external sources that had to do with weather related events that had to do with political events. Social media feeds a separate I'm overly that on top off our map of interest with regard to our supply chain sites and also where they were supposed to deliver. We also leave them our capabilities here, track of those shipments as they flowed and have that data flow back as well so that we would know exactly where where things were. This is only possible because we had a streamline data ops capability and we have built this Central Data and AI platform for IBM. Now you flip over to the Coleman 19 situation when Corbyn 19 merged and we began to realize that this was going to be a significant significant pandemic. What we were able to do very quickly wants to overlay the over 19 incidents on top of our sites of interest, as well as pick up what was being reported about those sites of interests and provide that over to our business continuity. So this became an immediate exercise that we embark. But it wouldn't have been possible if you didn't have the foundation off the data office pipeline as well as that Central Data and AI platform even plays to help you do that very, very quickly and adapt. >>So what I really like about this story and something that I want to drill into is it Essentially, a lot of organizations have a really tough time operational izing ai, infusing it to use your word and the fact that you're doing it, um is really a good proof point that I want to explore a little bit. So you're essentially there was a number of aspects of what you just described. There was the data quality piece with your data quality in theory, anyway, is going to go up with more data if you can handle it and the other was speed time to insight, so you can respond more quickly if it's talk about this Covic situation. If you're days behind for weeks behind, which is not uncommon, sometimes even worse, you just can't respond. I mean, the things change daily? Um, sometimes, Certainly within the day. Um, so is that right? That's kind of the the business outcome. An objective that you guys were after. >>Yes, you know, So Rama Common infuse ai into your business processes right over our chain. Um, don't come metric. That one focuses on is end to end cycle time. So you take that process the end to end process and you're trying to reduce the end to end cycle time by several factors, several orders of magnitude. And you know, there are some examples off things that we did. For instance, in my organ organization that has to do with the generation of metadata is data about data. And that's usually a very time consuming process. And we've reduced that by over 95%. By using AI, you actually help in the metadata generation itself. And that's applied now across the board for many different business processes that, you know IBM has. That's the same kind of principle that was you. You'll be able to do that so that foundation essentially enables you to go after that cycle time reduction right off the bat. So when you get to a situation like over 19 situation which demands urgent action. Your foundation is already geared to deliver on that. >>So I think actually, we might have a graphic. And then the second graphic, guys, if you bring up a 2nd 1 I think this is Interpol. What you're talking about here, that sort of 95% reduction. Ah, guys, if you could bring that up, would take a look at it. So, um, this is maybe not a cove. It use case? Yeah. Here it is. So that 95% reduction in the cycle time improvement in data quality. What we talked about this actually some productivity metrics, right? This is what you're talking about here in this metadata example. Correct? >>Yeah. Yes, the metadata. Right. It's so central to everything that one does with. I mean, it's basically data about data, and this is really the business metadata that you're talking about, which is once you have data in your data lake. If you don't have business metadata describing what that data is, then it's very hard for people who are trying to do things to determine whether they can, even whether they even have access to the right data. And typically this process is being done manually because somebody looks at the data that looks at the fields and describe it. And it could easily take months. And what we did was we essentially use a deep learning and natural language processing of road. Look at all the data that we've had historically over an idea, and we've automated metadata generation. So whether it was, you know, you were talking about the data relevant for 19 or for supply chain or far receivable process any one of our business processes. This is one of those fundamental steps that one must go through. You'll be able to get your data ready for action. And if you were able to take that cycle time for that step and reduce it by 95% you can imagine the acceleration. >>Yeah, and I like you were saying before you talk about the end to end concept, you're applying system thinking here, which is very, very important because, you know, a lot of a lot of clients that I talk to, they're so focused on one metric maybe optimizing one component of that end to end, but it's really the overall outcome that you're trying to achieve. You may sometimes, you know, be optimizing one piece, but not the whole. So that systems thinking is very, very important, isn't it? >>The systems thinking is extremely important overall, no matter you know where you're involved in the process off designing the system. But if you're the data guy, it's incredibly important because not only does that give you an insight into the cycle time reduction, but it also give clues U N into what standardization is necessary in the data so that you're able to support an eventual out. You know, a lot of people will go down the part of data governance and the creation of data standards, and you can easily boil the ocean trying to do that. But if you actually start with an end to end, view off your key processes and that by extension the outcomes associated with those processes as well as the user experience at the end of those processes and kind of then work backwards as one of the standards that you need for the data that's going to feed into all that, that's how you arrive at, you know, a viable practical data standards effort that you can essentially push forward so that there are multiple aspect when you take that end to end system view that helps the chief legal. >>One of the other tenants of data ops is really the ability across the organization for everybody to have visibility. Communications is very key. We've got another graphic that I want to show around the organizational, you know, in the right regime, and it's a complicated situation for a lot of people. But it's imperative, guys, if you bring up the first graphic, it's a heritage that organizations, you know, find bringing the right stakeholders and actually identify those individuals that are going to participate so that this full visibility everybody understands what their roles are. They're not in silos. So, guys, if you could show us that first graphic, that would be great. But talk about the organization and the right regime there. Interpol? >>Yes, yes, I believe you're going to know what you're going to show up is actually my organization, but I think it's yes, it's very, very illustrative what one has to set up. You'll be able to pull off the kind of impact that I thought So let's say we talked about that Central Data and AI platform that's driving the entire enterprise, and you're infusing AI into key business processes like the supply chain. Then create applications like the operational risk in size that we talked about that extended over. Do a fast emerging and changing situation like the over 19. You need an organization that obviously reflects the technical aspects of the right, so you have to have the data engineering on and AI on. You know, in my case, there's a lot of emphasis around deep learning because that's one of those skill set areas that's really quite rare, and it also very, very powerful. So uh huh you know, the major technology arms off that. There's also the governance on that I talked about. You have to produce the set off standards and implement them and enforce them so that you're able to make this into an impact. But then there's also there's a there's an adoption there. There's a There's a group that reports into me very, very, you know, Empowered Group, which essentially has to convince the rest of the organization to adopt. Yeah, yeah, but the key to their success has been in power in the sense that they're on power. You find like minded individuals in our key business processes. We're also empowered. And if they agree that just move forward and go and do it because you know, we've already provided the central capabilities by Central. I don't mean they're all in one location. You're completely global and you know it's it's It's a hybrid multi cloud set up, but it's a central in the sense that it's one source to come for for trusted data as well as the the expertise that you need from an AI standpoint to be able to move forward and deliver the business out. So when these business teams come together, be an option, that's where the magic happens. So that's another another aspect of the organization that's critical. And then we've also got, ah, Data Officer Council that I chair, and that has to do with no people who are the chief data officers off the individual business units that we have. And they're kind of my extended teams into the rest of the organization, and we levers that bolt from a adoption off the platform standpoint. But also in terms of defining and enforcing standards. It helps them stupid. >>I want to come back over and talk a little bit about business resiliency people. I think it probably seen the news that IBM providing supercomputer resource is that the government to fight Corona virus. You've also just announced that that some some RTP folks, um, are helping first responders and non profits and providing capabilities for no charge, which is awesome. I mean, it's the kind of thing. Look, I'm sensitive companies like IBM. You know, you don't want to appear to be ambulance chasing in these times. However, IBM and other big tech companies you're in a position to help, and that's what you're doing here. So maybe you could talk a little bit about what you're doing in this regard. Um, and then we'll tie it up with just business resiliency and importance of data. >>Right? Right. So, you know, I explained that the operational risk insights application that we had, which we were using internally, we call that 19 even we're using. We're using it primarily to assess the risks to our supply chain from various events and then essentially react very, very quickly. Do those doodles events so you could manage the situation. Well, we realize that this is something that you know, several non government NGOs that they could essentially use. There's a stability because they have to manage many of these situations like natural disaster. And so we've given that same capability, do the NGOs to you and, uh, to help that, to help them streamline their planning. And there's thinking, by the same token, But you talked about over 19 that same capability with the moment 19 data over layed on double, essentially becomes a business continuity, planning and resilience. Because let's say I'm a supply chain offers right now. I can look at incidents off over night, and I can I know what my suppliers are and I can see the incidents and I can say, Oh, yes, no, this supplier and I can see that the incidences going up this is likely to be affected. Let me move ahead and stop making plans backup plans, just in case it reaches a crisis level. On the other hand, if you're somebody in revenue planning, you know, on the finance side and you know where you keep clients and customers are located again by having that information over laid that those sites, you can make your own judgments and you can make your own assessment to do that. So that's how it translates over into business continuity and resolute resilience planning. True, we are internally. No doing that now to every department. You know, that's something that we're actually providing them this capability because we build rapidly on what we have already done to be able to do that as we get inside into what each of those departments do with that data. Because, you know, once they see that data, once they overlay it with their sights of interest. And this is, you know, anybody and everybody in IBM, because no matter what department they're in, there are going to decide the interests that are going to be affected. And they haven't understanding what those sites of interest mean in the context off the planning that they're doing and so they'll be able to make judgments. But as we get a better understanding of that, we will automate those capabilities more and more for each of those specific areas. And now you're talking about the comprehensive approach and AI approach to business continuity and resilience planning in the context of a large IT organization like IBM, which obviously will be of great interest to our enterprise, clients and customers. >>Right? One of the things that we're researching now is trying to understand. You know, what about this? Prices is going to be permanent. Some things won't be, but we think many things will be. There's a lot of learnings. Do you think that organizations will rethink business resiliency in this context that they might sub optimize profitability, for example, to be more prepared crises like this with better business resiliency? And what role would data play in that? >>So, you know, it's a very good question and timely fashion, Dave. So I mean, clearly, people have understood that with regard to that's such a pandemic. Um, the first line of defense, right is is not going to be so much on the medicine side because the vaccine is not even available and will be available for a period of time. It has to go through. So the first line of defense is actually think part of being like approach, like we've seen play out across the world and then that in effect results in an impact on the business, right in the economic climate and on the business is there's an impact. I think people have realized this now they will honestly factor this in and do that in to how they do become. One of those things from this is that I'm talking about how this becomes a permanent. I think it's going to become one of those things that if you go responsible enterprise, you are going to be landing forward. You're going to know how to implement this, the on the second go round. So obviously you put those frameworks and structures in place and there will be a certain costs associated with them, and one could argue that that would eat into the profitability. On the other hand, what I would say is because these two points really that these are fast emerging fluid situations. You have to respond very, very quickly. You will end up laying out a foundation pretty much like we did, which enables you to really accelerate your pipeline, right? So the data ops pipelines we talked about, there's a lot of automation so that you can react very quickly, you know, data injection very, very rapidly that you're able to do that kind of thing, that meta data generation. That's the entire pipeline that you're talking about, that you're able to respond very quickly, bring in new data and then aggregated at the right levels, infuse it into the work flows on the delivery, do the right people at the right time. Well, you know that will become a must. But once you do that, you could argue that there's a cost associated with doing that. But we know that the cycle time reductions on things like that they can run, you know? I mean, I gave you the example of 95% 0 you know, on average, we see, like a 70% end to end cycle time where we've implemented the approach, and that's been pretty pervasive within IBM across the business. So that, in essence, then actually becomes a driver for profitability. So yes, it might. You know this might back people into doing that, but I would argue that that's probably something that's going to be very good long term for the enterprises and world, and they'll be able to leverage that in their in their business and I think that just the competitive director off having to do that will force everybody down that path. But I think it'll be eventually ago >>that end and cycle time. Compression is huge, and I like what you're saying because it's it's not just a reduction in the expected loss during of prices. There's other residual benefits to the organization. Interpol. Thanks so much for coming on the Cube and sharing this really interesting and deep case study. I know there's a lot more information out there, so really appreciate your done. >>My pleasure. >>Alright, take everybody. Thanks for watching. And this is Dave Volante for the Cube. And we will see you next time. Yeah, yeah, yeah.

Hi, I'm Peter Burris, and this is Wikibon's Action Item. We're joined here in the studio by David Floyer. Hi David. >> Hi there. >> And remote, we've got Jim Kobielus. Hi, Jim. >> Hi everybody. >> Now, Jim, you probably can't see this, but for those who are watching, when we do see the broad set, notice that David Floyer's got his Game of Thrones coffee cup with us. Now that has nothing to do with the topic. David, and Jim, we're going to be talking about this challenge that businesses have, that enterprises have, as they think about making practical use of AI. The presumption for many years was that we were going to move all the data up into the Cloud in a central location, and all workloads were going to be run there. As we've gained experience, it's very clear that we're actually going to see a greater distribution function, partly in response to a greater distribution of data. But what does that tell about the relationship between AI, AI workloads, storage, and hybrid Cloud? David, why don't you give us a little clue as to where we're going to go from here. >> Well I think the first thing we have to do is separate out the two types of workload. There's the development of the AI solution, the inference code, et cetera, the dealing with all of the data required for that. And then there is the execution of that code, which is the inference code itself. And the two are very different in characteristics. For the development, you've got a lot of data. It's very likely to be data-bound. And storage is a very important component of that, as well as computer and the GPUs. For the inference, that's much more compute-bound. Again, compute neural networks, GPUs, are very, very relevant to that portion. Storage is much more ephemeral in the sense that the data will come in and you will need to execute on it. But that data will be part of the, the compute will be part of that sensor, and you will want the storage to be actually in the DIMM itself, or non-volatile DIMM, right up as part of the processing. And you'll want to share that data only locally in real time, through some sort of mesh computing. So, very different compute requirements, storage requirements, and architectural requirements. >> Yeah, let's go back to that notion of the different storage types in a second, but Jim, David described how the workloads are going to play out. Give a sense of what the pipelines are going to look like, because that's what people are building right now, is the pipelines for actually executing these workloads. How will they differ? How do they differ in the different locations? >> Yeah, so the entire DataOps pipeline for data science, data analytics, AI in other words. And so what you're looking at here is all the processes from discovering and adjusting the data to transforming and preparing and correcting it, cleansing it, to modeling and training the AI models, to serving them out for inferencing along the lines of what David's describing. So, there's different types of AI models and one builds from different data to do different types of inferencing. And each of these different pipelines might be highly, often is, highly specific to a particular use case. You know, AI for robotics, that's a very different use case from AI for natural language processing, embedded for example in an e-commerce portal environment. So, what you're looking at here is different pipelines that all share a common sort of flow of activities and phases. And you need a data scientist to build and test, train and evaluate and serve out the various models to the consuming end devices or application. >> So, David we've got 50 or so years of computing. Where the primary role of storage was to assist a transaction and the data associated with that transaction that has occurred. And that's you know, disk and then you have all the way out to tape if we're talking about archive. Flash changes that equation. >> Absolutely changes it. >> AI absolutely demands a different way of thinking. Here we're not talking about persisting our data we're talking about delivering data, really fast. As you said, sometimes very ephemeral. And so, it requires a different set of technologies. What are some of the limitations that historically storage has been putting on some of these workloads? And how are we breaching those limitations, to make them possible? >> Well if we take only 10 years ago, the start of the big data was Hadoop. And that was spreading the data over very cheap disks and hard disks. With the compute there, and you spread that data and you did it all in parallel on very cheap nodes. So, that was the initial but that is a very expensive way of doing it now because you're tying the data to that set of nodes. They're all connected together so, a more modern way of doing it is to use Flash, to use multiple copies of that data but logical copies or snapshots of that Flash. And to be able to apply as many processes, nodes as is appropriate for that particular workload. And that is a far more efficient and faster way of processing that or getting through that sort of workload. And it really does make a difference of tenfold in terms of elapsed time and ability to get through that. And the overall cost is very similar. >> So that's true in the inferencing or, I'm sorry, in the modeling. What about in the inferencing side of things? >> Well, the inferencing side is again, very different. Because you are dealing with the data coming in from the sensors or coming in from other sensors or smart sensors. So, what you want to do there is process that data with the inference code as quickly as you can, in real time. Most of the time in real time. So, when you're doing that, you're holding the current data actually in memory. Or maybe in what's called non-volatile DIMM and VDIMM. Which gives you a larger amount. But, you almost certainly don't have the time to go and store that data and you certainly don't want to store it if you can avoid it because it is a large amount of data and if I open my... >> Has limited derivative use. >> Exactly. >> Yeah. >> So you want to get all or quickly get all the value out of that data. Compact it right down using whatever techniques you can, and then take just the results of that inference up to other ones. Now at the beginning of the cycle, you may need more but at the end of the cycle, you'll need very little. >> So Jim, the AI world has built algorithms over many, many, many years. Many which still persist today but they were building these algorithms with the idea that they were going to use kind of slower technologies. How is the AI world rethinking algorithms, architectures, pipelines, use cases? As a consequence of these new storage capabilities that David's describing? >> Well yeah, well, AI has become widely distributed in terms of its architecture increasingly and often. Increasingly it's running over containerized, Kubernetes orchestrated fabrics. And a lot of this is going on in the area of training, of models and distributing pieces of those models out to various nodes within an edge architecture. It may not be edge in the internet of things sense but, widely distributed, highly parallel environments. As a way of speeding up the training and speeding up the modeling and really speeding up the evaluation of many models running in parallel in an approach called ensemble modeling. To be able to converge on a predictive solution, more rapidly. So, that's very much what David's describing is that that's leveraging the fact that memory is far faster than any storage technology we have out there. And so, being able to distribute pieces of the overall modeling and training and even data prep of workloads. It's able to speed up the deployment of highly optimized and highly sophisticated AI models for the cutting edge, you know, challenges we face like the Event Horizon telescope for example. That we're all aware of when they were able to essentially make a visualization of a black hole. That relied on a form of highly distributed AI called Grid Computing. For example, I mean the challenges like that demand a highly distributed memory-centric orchestrated approach to tackling. >> So, you're essentially moving the code to the data as opposed to moving all of the data all the way out to the one central point. >> Well so if we think about that notion of moving code to the data. And I started off by suggesting that. In many respects, the Cloud is an architectural approach to how you distribute your workloads as opposed to an approach to centralizing everything in some public Cloud. I think increasingly, application architects and IT organizations and service providers are all seeing things in that way. This is a way of more broadly distributing workloads. Now as we think about, we talked briefly about the relationship between storage and AI workloads but we don't want to leave anyone with the impression that we're at a device level. We're really talking about a network of data that has to be associated with a network of storage. >> Yes. >> Now that suggests a different way of thinking about how - about data and data administration storage. We're not thinking about devices, we're really trying to move that conversation up into data services. What kind of data services are especially crucial to supporting some of these distributed AI workloads? >> Yes. So there are the standard ones that you need for all data which is the backup and safety and encryption security, control. >> Primary storage allocation. >> All of that, you need that in place. But on top of that, you need other things as well. Because you need to understand the mesh, the distributed hybrid Cloud that you have, and you need to know what the capabilities are of each of those nodes, you need to know the latencies between each of those nodes - >> Let me stop you here for a second. When you say "you need to know," do you mean "I as an individual need to know" or "the system needs to know"? >> It needs to be known, and it's too complex, far too complex for an individual ever to solve problems like this so it needs, in fact, its own little AI environment to be able to optimize and check the SLAs so that particular inference coding can be achieved in the way that it's set up. >> So it sounds like - >> It's a mesh type of computer. >> Yeah, so it sounds like one of the first use cases for AI, practical, commercial use cases, will be AI within the data plane itself because the AI workloads are going to drive such a complex model and utilization of data that if you don't have that the whole thing will probably just fold in on itself. Jim, how would you characterize this relationship between AI inside the system, and how should people think about that and is that really going to be a practical, near-term commercial application that folks should be paying attention to? >> Well looking at the Cloud native world, what we need and what we're increasingly seeing out there are solutions, tools, really data planes, that are able to associate a distributed storage infrastructure of a very hybridized nature in terms of disk and flash and so forth with a highly distributed containerized application environment. So for example just last week at Jeredhad I met with the folks from Robin Systems and they're one of the solution providers providing those capabilities to associate, like I said, the storage Cloud with the containerized, essentially application, or Cloud applications that are out there, you know, what we need there, like you've indicated, are the ability to use AI to continue to look for patterns of performance issues, bottlenecks, and so forth and to drive the ongoing placement of data storage nodes and servers which in clusters and so forth as way of making sure that storage resources are always used efficiently that SLAs as David indicated are always observed in an automated fashion as the native placement and workload placement decisions are being made and so ultimately that the AI itself, whatever it's doing like recognizing faces or recognizing human language, is able to do it as efficiently and really as cheaply as possible. >> Right, so let me summarize what we've got so far. We've got that there is a relationship between storage and AI, that the workload suggests that we're going to have centralized modeling, large volumes of data, we're going to have distributed inferencing, smaller on data, more complex computing. Flash is crucial, mesh is crucial, and increasingly because of the distributed nature of these applications, there's going to have to be very specific and specialized AI in the infrastructure, in that mesh itself, to administer a lot of these data resources. >> Absolutely. >> So, but we want to be careful here, right David? We don't want to suggest that we have, just as the notion of everything goes into a centralized Cloud under a central administrative effort, we also don't want to suggest this notion that there's this broad, heterogeneous, common, democratized, every service available everywhere. Let's bring hybrid Cloud into this. >> Right. >> How will hybrid Cloud ultimately evolve to ensure that we get common services where we need them? And know where we don't have common services so that we can factor those constraints? >> So it's useful to think about the hybrid Cloud from the point of view of the development which will be fairly normal types of computing and be in really large centers and the edges themselves, which will be what we call autonomous Clouds. Those are the ones at the edge which need to be self-sufficient. So if you have an autonomous car, you can't guarantee that you will have communication to it. And most - a lot of IOTs in distant places which again, on chips or distant places, where you can't guarantee. So they have to be able to run much more by themselves. So that's one important characteristic so that autonomous one needs to be self-sufficient itself and have within it all the capabilities of running that particular code. And then passing up data when it can. >> Now you gave examples where it's physically required to do that, but it's also OT examples. >> Exactly. >> Operational technologies where you need to have that air gap to ensure that bad guys can't get into your data. >> Yes, absolutely, I mean if you think about a boat, a ship, it has multiple very clear air gaps and a nuclear power station has a total air gap around it. You must have those sort of air gaps. So it's a different architecture for different uses for different areas. But of course data is going to come up from those autonomous, upwards, but it will be a very small amount of the data that's actually being processed. The data, and there'll be requests down to those autonomous Clouds for additional processing of one sort or another. So there still will be a discussion, communication, between them, to ensure that the final outcome, the business outcome, is met. >> All right, so I'm going to ask each of you guys to give me a quick prediction. David, I'm going to ask you about storage and then Jim I'm going to ask you about AI in light of David's prediction about storage. So David, as we think about where these AI workloads seem to be going, how is storage technology going to evolve to make AI applications easier to deal with, easier to run, cheaper to run, more secure? >> Well, the fundamental move is towards larger amounts of Flash. And the new thing is that larger amounts of non-volatile DIMM, the memory in the computer itself, those are going to get much, much bigger, those are going to help with the execution of these real-time applications and there's going to be high-speed communication between short distances between the different nodes and this mesh architecture. So that's on the inference side, there's a big change happening in that space. On the development side the storage will move towards sharing data. So having a copy of the data which is available to everybody, and that data will be distributed. So sharing that data, having that data distributed, will then enable the sorts of ways of using that data which will retain context, which is incredibly important, and avoid the cost and the loss of value because of the time taken of moving that data from A to B. >> All right, so to summarize, we've got a new level in the storage hierarchy that puts between Flash and memory to really accelerate things, and then secondly we've got this notion that increasingly we have to provide a way of handling time and context so that we sustain fidelity especially in more real-time applications. Jim, given that this is where storage is going to go, what does that say about AI? >> What it says about AI is that first of all, we're talking about like David said, meshes of meshes, every edge node is increasingly becoming a mesh in its own right with disparate CPUs and GPUs and whatever, doing different inferencing on each device, but every one of these, like a smart car, will have plenty of embedded storage to process a lot of data locally that may need to be kept locally for lots of very good reasons, like a black box in case of an accident, but also in terms of e-discovery of the data and the models that might have led up to an accident that might have caused fatalities and whatnot. So when we look at where AI is going, AI is going into the mesh of mesh, meshes of meshes, where there's AI running it in each of the nodes within the meshes, and the meshes themselves will operate as autonomous decisioning nodes within a broader environment. Now in terms of the context, the context increasingly that surrounds all of the AI within these distributed architectures will be in the form of graphs and graphs are something distinct from the statistical algorithms that we built AI out of. We're talking about knowledge graphs, we're talking about social graphs, we're talking about behavioral graphs, so graph technology is just getting going. For example, Microsoft recently built, they made a big continued push into threading graph - contextual graph technology - into everything they do. So that's where I see AI going is up from statistical models to graph models as the broader metadata framework for binding everything together. >> Excellent. All right guys, so Jim, I think another topic another time might be the mesh mess. (laughs) But we won't do that now. All right, let's summarize really quickly. We've talked about how the relationship between AI, storage and hybrid Clouds are going to evolve. Number one, AI workloads are at least differentiated by where we handle modeling, large amounts of data still need a lot of compute, but we're really focused on large amounts of data and moving that data around very, very quickly. But therefore proximate to where the workload resides. Great, great application for Clouds, large, public as well as private. On the other side, where the inferencing work is done, that's going to be very compute-bound, smaller data volumes, but very, very fast data. Lot of flash everywhere. The second thing we observed is that these new AI applications are going to be used and applied in a lot of different domains, both within human interaction as well as real-time domains within IOT, et cetera, but that as we evolve, we're going to see a greater relationship between the nature of the workload and the class of the storage, and that is going to be a crucial feature for storage administrators and storage vendors over the next few year is to ensure that that specialization is reflected in what's known. What's needed. Now the last point that we'll make very quickly is that as we look forward, the whole concept of hybrid Cloud where we can have greater predictability into the nature of data-oriented services that are available for different workloads is going to be really, really important. We're not going to have all data services common in all places. But we do want to make sure that we can assure whether it's a container-based application or some other structure, that we can ensure that the data that is required will be there in the context, form and metadata structures that are required. Ultimately, as we look forward, we see new classes of storage evolving that bring data even closer to the compute side, and we see new data models emerging, such as graph models, that are a better overall reflection of how this distributed data is going to evolve within hybrid Cloud environments. David Floyer, Jim Kobielus, Wikibon analysts, I'm Peter Burris, once again, this has been Action Item.

(electronic music) >> Hi, I'm Peter Burris. Welcome to Wikibon action item from theCUBE Studios in Palo Alto, California. So today we've got a great conversation and what we're going to be talking about is hybrid cloud. Hybrid cloud's been in the news a lot, lately. Larger consequences, from changes made by AWS, as they announced Outpost, and acknowledged for the first time that there's going to be a greater distribution of data and a greater distribution of function as enterprises, move to the cloud. We've been on top of this for quite some time, and have actually coined what we called true hybrid cloud. Which is the idea that increasingly we're going to see a need for a common set of capabilities and services, in multiple locations, so that the cloud can move to the data, and not the data automatically being presumed to move to the cloud. Now to have that conversation and to reveal some new research on what the cost and value propositions of the different options are that are available today. We've got David Foyer, David welcome to theCUBE. >> Thank you. >> So David, let's start, when we talk about hybrid cloud, we are seeing, a continuum of different options starting to emerge. What are the defining characteristics? >> So, yes, we're seeing a continuum emerging. We have a what we call stand alone of course at one end of the spectrum, and then we have multi cloud, and then we have loosely and tightly coupled, and then we have true, and as you go up the spectrum. So the dependents upon data depend on the data plain, dependents upon low latency, dependents on writing a systems of record, records. All of those increase as we're going from high latency and high bandwidth all the way up to low latency. >> So let me see if I got that right. So true hybrid cloud is at one end. >> Yes. >> And true hybrid cloud is, low latency, right on your work loads, simple as possible administration. That means we are typically going to have, a common stack in all locations. >> Yes. >> Next to that is this notion of tightly coupled, hybrid cloud, which could be higher latency, write oriented, could probably has a common set of software, on all nodes, that handles state. And then kind of this notion of loosely coupled. Multi well hybrid cloud, which is high latency, read oriented, which may have just API level coordination and commonality on all nodes. >> Yep that's right, and then you go down even further to just multi cloud, where you're just connecting things and each of them is independent of each other. >> So if I'm a CIO and I'm looking at a move to a cloud, I have to think about green field applications and the natural distribution of data for those green field applications, and that's going to help me choose which class of hybrid cloud, I'm going to use. But let's talk about the more challenging set of scenarios for most CIO's, which is the existing legacy applications. >> The systems of record. >> Yeah, the systems of record as I try to bring those, those cloud like experience to those applications, how am I going through that thought process? >> So, we have some choices, the choices are I could move it up too lift and shift, up to one of the cloud's, one of the large cloud's, many of them are around, and what if I, if I do that, what I need to be looking at is, what is the cost of moving that data, and what is the cost of pushing that up into the cloud, and what's the conversion cost, if I needed to move, to another database. >> And I think that's the biggest one, so that's just the cost of moving the data, which is just an ingress cost, it's a cost of format changes. >> Absolutely >> You know, migration and all the other elements, conversion changes et cetera. >> Right, so what I did in my research was focus on systems of record, the highly expensive, very, very important systems of record, which obviously are fed by a lot of other things, you know, systems of engagements, analytics et cetera. But those systems of record have to work. You need to know if you've taken an order. You need to have consistency about that order. You need to know always that you can recover any data, you need in your financials et cetera, all of that is mission critical systems of record. And that's the piece that I focused on here, and I focused on. >> So again these are low latency. >> Very, low latency, yes. >> Write oriented. >> Very write oriented, types of applications, and, I focused on Oracle because the majority, of systems of record, run on Oracle databases, the large scale ones at least. So, that's what we are focusing on here. So I, looking at the different options for a CIO, of how they would go, and there are three main options open at the moment, there's Oracle, Cloud, Cloud at customer, which gives the cloud experience. There is Microsoft Azure Stack, which has a Oracle database version of it, and Outposts, but we eliminated Outposts not because, it's not going to be any good, but because it's not there yet. >> You can't do research on it if it doesn't exist yet. >> (laughs) That's right. So, we focused on Oracle and Azure, and we focused on, what was the benefit of moving from a traditional environment, where you've got best of breed essentially on site, to this cloud environment. >> So, if we think about it, the normal way of thinking about this kind of a research, is that people talk about R.O.I, and historically that's been done by looking, by keeping the amount of work that's performed, as given, constant and then looking at how the different technology components compare from a call standpoint. But a move to Cloud, the promise of a move to Cloud is not predicated on lowering costs per say. You may have other financial considerations of course but, it's really predicated on the notion of the cloud experience. Which is intended to improve business results, so if we think about R.O.I, as being a numerator question, with the value is the amount of work you do, versus a denominator question which is what resources are consumed to perform that work. It's not just the denominator side, we really need to think about the numerator side as well. >> The value you are creating, yes. >> So, what kind of thing's are we focused on when we think about that value created, as a consequence of possibilities and options of the Cloud. >> Right, so both are important, so obviously when you move, to a cloud environment, you can simplify operations in particular, you can simplify recovery, you can simplify a whole number of things within the IT shop and those give you extra resources. And then the question is, do you just cash in on those resources and say okay I've made some changes, or do you use those resources to improve the ability of your systems to work. One important characteristic of IT, all IT and systems of record in particular, is that you get depreciation of that asset. Over time it becomes less fitted, to the environment, that it started with, so you have to do maintenance on it. You have to do maintenance and work, and as you know, most work done in an IT shop is on the maintenance side. >> Meaning it's an enhancement. >> It's maintenance and enhancement, yes. So making more resources available, and making it easier to do that maintenance, and making less things that are going to interfere with that, faster time to maintenance, faster time to new applications or improvements. Is really fundamental to systems of record. So that is the value that you can bring to it and you also bring value with lower better availability, higher availability as well. So those are the thing's we have put into the model, to see how the different approaches, and we were looking at really a total, one supplier being responsible for everything, which was the Oracle environment, and Oracle Cloud at Customer, to a sort of hybrid environment, more hybrid environment, where you had. >> Or mixed, or mixed. >> Mixed environment, yes. Where you had the equipment coming from different places. >> One vendor. >> The service, the Oracle, the Azure service, coming from Microsoft, and of course the database coming then from Oracle itself. And we found tremendous improvement in the value that you could get because of the single source. We found that a better model. >> So, the common source lead to efficiencies, that then allowed a business to generate new classes, of value >> Correct. >> Cause' as you said, you know, 70 plus percent of an IT or business is spent on technology, is associated with maintaining what's there, enhancing what's there, and a very limited amount is focused on new green field, and new types of applications. So if you can reduce the amount of time and energy, that goes into that heritage set of applications, those systems of record, then that opens up, that frees up resources to do some other things. >> And, having the flexibility now with things like Azure Stack and in the future AWS, of putting that resource either on premise or in the cloud, means that you can make decisions about where you process these things, about where the data is, about where the data needs to be, the best placement for the data for what you're trying to do. >> That decision is predicated on things like latency, but also regulatory environment, intellectual property control. >> And the cost of moving data up and down. So the three laws of the cloud, so having that flexibility of keeping it where you want to is a tremendous value in again, in terms of, the speed of deployment and the speed of improvement. >> So we'll get to the issues surrounding the denominator side of this. I want to come back to that numerator side. So the denominator again is, the resources consume, to deliver the work to the business, but when we talk about that denominator side, perhaps opening up additional monies, to do new types of development, new types of work. But, take us through some of the issues like what is a cloud experience associated with, single vendor, faster development, give us some of the issues that are really driving the value proposition above the line. >> The whole issue about Cloud is that you take away all of the requirements to deal with the hardware, deal with the orchestration of the storage, deal with all of these things, so instead of taking weeks, months to put in extra resources, you say, I want them, and it's there. >> So you're taking administrative tasks, out of the flow. >> Out of the flow yes. >> And as a consequence, things happen faster, so time of value is one of the first ones, give us another one. >> So, obviously the ability to have. It's a cloud environment, so if you're a vendor, of that cloud, what you want to be able to do, is to make incremental changes, quickly as opposed to waiting for a new release and work on the release basis. So that fundamental speed to change, speed to improve, bring in new features, bring in new services, a cloud first type model, that is a very powerful way for the vendor to push out new things, and for the consumer to absorb them. >> Right, so the first one is time to value, but also it's lower cost to innovation. >> Yes, faster innovation, ability to innovate. And then the third most important part is, if you re-invest those resources that you have saved into new services, new capabilities of doing that, to me the most important thing, long term for systems of record is to be able to make them go faster, and use that extra latency time there to bring in systems of analytics, AI systems, other systems, and provide automation of individual business processes, increased automation. That is going to happen overtime, that's a slow adding to it, but it means you can use those cloud mechanisms, those additional resources, wherever they are. You can use those to provide, a clear path to, improving the current systems of record. And that is a more faster and more cost effective way, than going in for a conversion, or moving the data up to the cloud or lift and shift, for these types of applications. >> So these are all kind of related, so I get superior innovation speeds, because I'm taking new technology and faster, I get faster time to value, because, I'm not having to perform a bunch of tasks, and I can get, I can imbue additional types of work, in support of automation, without dramatically, expanding the transactional latency and a rival way of transactions within the system of record. Okay so, how did Oracle and Azure, with Oracle, stack up in your analysis? >> So first of all important is both are viable solutions, they both would work. Okay, but the impact in terms of the total business value including obviously any savings on people and things like that, was 290 nearly, $300 million additional, this was for a >> For how big a company? >> For a fortune 2000 customer, so it was around two billion dollars, so a lot of money, over five years, a lot of money. Either way, you would save 200 million, if you were with the Azure, but 300 with the Oracle. So that to me is far, far higher than the costs of IT, for that particular company, it's a strategic decision, to be able to get more value out quicker, and for this class of work load, on Oracle, then Oracle at Cloud was the best decision, to be absolutely fair, if you were on Microsoft's database, and you wanted to go to Microsoft Azure, that would be the better bet. You would get back a lot of those benefits. >> So stay within the stack if you can. >> Correct. >> Alright, so, two billion dollars a year, five years. $10 billion revenue, roughly. >> Between 200 million in saving, for One Microsoft, Azure, plus Oracle. 300 million so a 1% swing, talk to us about speed, value what happens in, a numerator side of that equation? >> So, it is lower in cost, but you have a higher, the cost of the actual cloud, is a little higher, so, overall the pure hardware, equipment class is a wash, it's not going to change much. >> Got it. >> It might be a little bit more expensive. You make the savings, as well because of the people, less operators, simpler environment. Those are the savings you're going to make, and then you are going to push those back, into the organization, as increase value that can be given to the line of the business. >> So the conclusion to the research is that if you are a CIO, you look at your legacy application, it's going to be difficult to move and you go with the stack that's best for those, legacy applications. >> Correct. >> So the vast majority of systems of record, are running on Oracle. >> Large scale. >> Large scale, then that means Oracle Cloud at Customers, is the superior fit for most circumstances. >> For a lot of those. >> If you're not there though then look at other options. >> Absolutely. >> Alright, David Foyer. >> Thank you. >> Thanks, very much for being on the cube today. And you've been watching another Wikibon action item, from theCUBE Studios in Palo Alto California, I'm Peter Burris, thanks very much for watching. (electronic music)

>> Hi, I'm Peter Burroughs Wellcome to wicked bon action >> item from the Cube Studios in Palo Alto, California So today we've got a great conversation, and what we're going to be talking about is hybrid cloud hybrid. Claude's been in the news a lot lately. Largest consequences from changes made by a Ws is they announced Outpost and acknowledged for the first time that there's going to be a greater distribution of data on a greater distribution of function as enterprise has moved to the cloud. We've been on top of this for quite some time, and it actually coined what we call true hybrid cloud, which is the idea that increasingly, we're going to see a need for a common set of capabilities and services in multiple locations so that the cloud could move to the data and not the data automatically being presumed to move to the cloud. >> Now to have that >> conversation and to reveal some new research on what the cost in value propositions of the different options are available. Today. We've >> got David Foyer. David. Welcome to the Cube. Thank you. So, David, let's start. When we talk about Hybrid Cloud, we're seeing a continuum of different options start to emerge. What are the defining characteristics? >> Yes, we're seeing it could continue him emerging. We have what we've called standalone off course. That one is end of the spectrum on DH. There we have multi cloud, and then we have loosely and tightly coupled, and then we have true and as you go up the spectrum. So the dependence upon data depends on the data plane dependence upon low latent see dependance on writing does a systems of record records. All of those increase as we going from from lonely for High Leighton Sea and High Band with all way up to low late. >> So let me see if I got this right. It's true. I've a cloud is at one end and true. Either cloud is low late and see right on into workloads simplest possible administration. That means we're typically goingto have a common stack in all locations. Next to that is this notion of tightly coupled hybrid cloud, which could be higher late. And see, right oriented could probably has a common set of software on all no common mental state. And then, kind of this. This notion of loosely coupled right multi or hybrid cloud, which is low, high late and see, write or read oriented, which may have just a P I level coordination and commonality and all >> that's right. And then you go down even further to just multi cloud, where you're just connecting things, and each of them is independent off each other. >> So if I'm a CEO and I'm looking at a move to a cloud, I have to think about Greenfield applications and the natural distribution of data for those Greenfield applications. And that's going to help me choose which class of hybrid clawed him and he used. But let's talk about the more challenging from a set of scenarios for most CEOs, which is the existing legacy applications as I cry that Rangel yeah, systems of record. As I try to bring those those cloud like experience to those applications, how am I going through that thought process? >> So we have some choices. The choices are I could move it up to lift and shift up to on a one of the clouds by the large clouds, many of them around. And what if I if I do that what I'm need to be looking at is, what is the cost of moving that data? And what is the cost of pushing that up into the cloud and lost the conversion cast if I need to move to another database, >> and I think that's the biggest one. So it just costs of moving the data, which is just uninterested. It's a cost of format changes at our migration and all the other out conversion changes. >> So what I did in my research was focus on systems of record, the the highly expensive, very, very important systems of record, which obviously are fed by a lot of other things their systems, the engagement analytics, etcetera. But those systems of record have to work. They you need to know if you've taken on order, you need to have consistency about that order. You need to know always that you can recover any data you need in your financials, etcetera. All of that is mission critical systems of record. Andi, that's the piece that I focused on here, and I focused on >> sort of. These are loaded and she >> low, very low, latent, right oriented, very right orientated types of applications. And I focused on the oracle because the majority ofthe systems of record run on Oracle databases on the large scale ones, at least so that's what we're we're focusing on here. So I looking at the different options for a C I O off. How they would go on DH. There are three main options open at the moment. There's there's Arkalyk Cloud Cloud, a customer, which gives thie the cloud experience. There is Microsoft as your stack, which has a a Oracle database version of it on DH outposts. But we eliminated outposts not because it's not going to be any good, but because it's not there yet, is >> you get your Razor John thing. >> That's right. So we focused on Oracle on DH as you and we focused on what was the benefit of moving from a traditional environment where you've got best of breed essentially on site to this cloud environment. >> So if we think about it, the normal way of thinking about this kind of a research is that people talk about R. A Y and historically that's been done by looking by keeping the amount of work that's performed has given constant and then looking at how the different technology components compare from a call standpoint. But a move to cloud the promise of a move to cloud is not predicated on lowering costs per se, but may have other financial considerations, of course, but it's really predicated on the notion of the cod experience, which is intended to improve business results. So we think about our lives being a numerator question. Value is the amount of work you do versus the denominator question, which is what resources are consumed to perform that work. It's not just the denominator side we really need to think about. The numerator side is well, you create. So what? What kind of things are we focused >> on? What we think about that value created his consequence of possibilities and options of the cloud. >> Right? So both are important. So so Obviously, when you move to a cloud environment, you can simplify operations. In particular, you can simplify recovery. You, Khun simplify a whole number of things within the shop and those give you extra resources on. Then the question is, Do you just cash in on those resources and say OK, I've made some changes, Or do you use those resources to improve the ability of your systems to work and one important characteristic off it alight and systems of record in particular is that you get depreciation of that asset. Over time, it becomes less fitted to the environment it has started with, so you have to do maintenance on it. You have to do maintenance and work, and as you know most means most work done in my tea shop is on the maintenance side minutes. An enhancement. It's maintenance. An enhancement, yes. So making more resources available on making it easier to do that maintenance are making less, less things that are going to interfere with that faster time to to to maintenance faster time. Two new applications or improvements is really fundamental to systems of record, so that is the value that you can bring to it. And you also bring value with lower of better availability, higher availability as well. So those are the things that we put into the model to see how the different approaches. And we were looking at really a total one. One supplier being responsible for everything, which was the Oracle environment of Oracle clouded customer to a sort of hybrid invite more hybrid environment where you had the the the work environment where you had the equipment coming from different place vendor that the service, the oracle, the as your service coming from Microsoft and, of course, the database coming then from Arkham itself. And we found from tremendous improvement in the value that you could get because of this single source. We found that a better model. >> So the common source led to efficiencies that then allowed a business to generate new classes of value. Because, as you said, you know, seventy plus percent of a night organ orb business is spending. Biology is associate with maintaining which they're enhancing. What's there in a very limited amount is focused on new greenfield or new types of applications. So if you can reduce the amount of time energy that goes into that heritage set of applications those systems of record, the not opens up that frees up resources to do some of the things >> on DH Having inflexibility now with things like As your stack conned in the future E. W. S off. Putting that resource either on premise or in the cloud, means that you can make decisions about where you process things things about where the data is about, where the data needs to be, the best placement of the data for what you're trying to do >> and that that decision is predicated on things like late in sea, but also regulatory, environment and intellectual property, controlling >> the custom moving data up and down. So the three laws of off off the cloud so having that flexibility of moving, keeping it where you want to, is a tremendous value in again in terms ofthe the speed of deployment on the speed of improved. >> So we'll get to the issues surrounding the denominator side of this. I want to come back to that numerator sites that the denominator again is the resources consumed to deliver the work to the business. But when we talk about that denominator side, know you perhaps opening up additional monies to do new types of development new times of work. But take us through some of the issues like you know what is a cloud experience associated with single vendor Faster development. Give us some of the issues that are really driving the value proposition. Look above the line. >> I mean, the whole issue about cloud is that you go on, take away all of the requirements to deal with the hardware deal with the orchestration off the storage deal with all of these things. So instead of taking weeks, months to put in extra resources, you say I want them on is there. >> So you're taking out administrate your taking administrative tasks out of the flow out of the flow, and as a consequence, things happen. Faster is the time of values. One of the first one. Give us another one. >> So obviously the ability to no I have it's a cloud environment. So if you're a vendor of that cloud, what you want to be able to do is to make incremental changes quickly, as opposed to awaiting for a new release and work on a release basis. So that fundamental speed to change speed to improve, bring in new features. Bringing new services a cloud first type model that is a very powerful way for the vendor to push out new things. And for the consumer, too, has absorbed them. >> Right? So the first one is time to value, but also it's lower cost to innovation. >> Yes, faster innovation ability to innovate. And then the third. The third most important part is if you if you re invest those resources that you've saved into new services new capabilities of doing that. To me, the most important thing long term for systems of record is to be able to make them go faster and use that extra Leighton see time there to bring in systems off systems of analytics A. I systems other systems on provide automation of individual business processes, increased automation that is gonna happen over time. That's that's a slow adding to it. But it means you can use those cloud mechanisms, those additional resources, wherever they are. You can use those to provide a clear path to improving the current systems of record. And that is a much faster and more cost effective way than going in for a conversion or moving the data upto the cloud or shifting lift and shift. For these types of acts, >> what kind of they're all kind of related? So I get, I get. I get superior innovation speeds because I'm taking new technology and faster. I get faster time to value because I'm not having to perform much of tasks, and I could get future could imbue additional types of work in support of automation without dramatically expanding the transactional wait and see on arrival rate of turns actions within the system of record. Okay, So how did Oracle and Azure with Oracle stack up in your analysis? >> So first of all, important is both a viable solutions. They both would work okay, but the impact in terms of the total business value, including obviously any savings on people and things like that, was two hundred nineteen eighty three hundred million dollars additional. This was for Robert to come in for a a Fortune two thousand customer, so it was around two billion dollars. So a lot of money over five years, a lot of money. Either way, you would save two hundred million if you were with with the zero but three hundred with the oracle, so that that to me, is far, far higher than the costs of I T. For that particular company, it's It is a strategic decision to be able to get more value out quicker. And for this class of workload on Oracle than Arkalyk, Cloud was the best decision to be absolutely fair If you were on Microsoft's database. And you wanted to go to Microsoft as you. That would be the better bet you would. You would get back a lot of those benefits, >> so stay with him. The stack, if you can't. Correct. All right, So So two billion dollars a year. Five years, ten billion dollars in revenue, roughly between two hundred million and saving for one Congress all around three. Treasure Quest. Oracle three hundred millions were one percent swing. Talk to us about speed value. What >> happens in the numerator side of that equation >> S Oh, so it is lower in caste, but you have a higher. The cast of the actual cloud is a little a little higher. So overall, the pure hardware equipment Cass is is awash is not going to change much. It might be a little bit more expensive. You make the savings a cz? Well, because of the people you less less operators, simpler environment. Those are the savings you're going to make. And then you're going to push those back into into the organization a cz increased value that could be given to the line of business. >> So the closure of the researchers If your CEO, you look at your legacy application going to be difficult to move, and you go with stack. That's best for those legacy applications. And since the vast majority of systems of record or running all scale large scale, then that means work. A cloud of customers is a superior fit for most from a lot of chances. So if you're not there, though, when you look at other options, all right, David Floy er thank you. Thanks very much for being on the Cube today, and you've been watching other wicked bon action >> item from the Cube Studios and Power Rialto, California on Peter Burke's Thanks very much for watching.

hi I'm Peter Burris and welcome to this week's action item once again we're broadcasting from our beautiful the cube Studios in Palo Alto California and the wiki bond team is a little bit smaller this week for variety of reasons I'm being joined remotely by Neil Raiden and Jim Kabila's how you doing guys we're doing great Peter I'd be good thank you alright and it's actually a good team what we're gonna talk about we're gonna be specifically talking about some interesting developments and 14 days or so gdpr is gonna kick in and people who are behind will find themselves potentially subject to significant fines we actually were talking to a chief privacy officer here in the US who told us that had the Equinix breach occurred in Europe after May 25 2008 eeen it would have cost or Equifax the Equifax breach it would have cost Equifax over 160 billion dollars so these are very very real types of money that we're talking about but as we started thinking about some of the implications of gdpr and when it's going to happen and the circumstances of its of its success or failure and what its gonna mean commercially to businesses we also started trying to fold in a second trend and that second trend is the role of bitcoins going to play Bitcoin has a number of different benefits we'll get into some of that in a bit but one of them is that the data is immutable and gdpr has certain expectations regarding a firm's flexibility and how it can manage and handle data and blockchain may not line up with some of those issues as well as a lot of the Braque blockchain advocates might think Jim what are some of the specifics well Peter yeah blockchain is the underlying distributed hyper ledger or trusted database underlying Bitcoin and many other things blockchain yeah you know the one of the core things about blockchain that makes it distinctive is that you can create records and append them to block change you can read from them but can't delete them or update them it's not a crud database it's essentially for you to be able to go in and you know and erase a personally identifiable information record on an EU subject is you EU citizen in a blockchain it's not possible if you stored it there in other words blockchain then at the very start because it's an immutable database would not allow you to comply with the GDP ours were quite that people have been given a right to be forgotten as what what it's called that is a huge issue that might put the big kibosh on implementation of blockchain not just for PII in the EU but really for multinational businesses anybody who does business in Europe and the core you know coordination is like you know we're disregard brexit for now like Germany and France and Italy you got to be conformant completely worldwide essentially with your in your your PII management capabilities in order to pass muster with the regulators in the EU and avoid these massive fines blockchain seems like it would be incompatible with that compliance so where does the blockchain industry go or does it go anywhere or will it shrink well the mania died because of the GDP our slap in the face probably not there is a second issue as well Jim Lise I think there is and that is blockchain is allows for anonymity which means that everybody effectively has a copy of the ledger anywhere in the world so if you've got personally identifiable information coming out of the EU and you're a member or you're a part of that blockchain Network living in California you get a copy of the ledger now you may not be able to read the details and maybe that protects folks who might implement applications in blockchain but it's a combination of both the fact that the ledger is fully distributed and that you can't go in and make adjustments so that people can be forgotten based on EU laws if I got that right that's right and then there's a gray area you can't encrypt any and every record in a blockchain and conceal it from the prying eyes of people in California or in Thailand or wherever in the EU but that doesn't delete it that's not the same as erasing or deleting so there's a gray issue and there's no clarity from the EU regulators on this what if you use secret keys to encrypt individual records PII on a blockchain and then lost the keys or deleted the keys is that effectively would that be the same as he racing the record even though those bits still be there to be unreadable none of this has really been addressed in practice and so it's all a gray area it's a huge risk factor for companies that are considering exploring uses of blockchain for managing identity and you know security and all that other good stuff related to the records of people living in EU member countries so it seems as though we have two things they're gonna have that are that are likely to happen first off it's very clear that a lot of the GDP are related regulations were written in advance of comprehending what blockchain might be and so it doesn't and GDP are typically doesn't dictate implementation styles so it may have to be amended to accommodate some of the blocks a blockchain implementation style but it also suggests that increasingly we're going to hear from a design standpoint the breaking up of data associated with a transaction so that some of the metadata associated with that transaction may end up in the blockchain but some of the actual PII related data that is more sensitive from a GDP or other standpoint might remain outside of the blockchain so the blockchain effectively becomes a distributed secure network for managing metadata in certain types of complex transactions this is is that is that in scope of what we're talking about Jim yeah I bet you've raised and alluded to a big issue for implementers there will be on chain implementations of particular data data applications and off chain implementations off chain off blockchain will probably be all the PII you know in databases relational and so forth that allow you to do deletes and updates and so forth in you know to comply with you know gdpr and so forth and similar mandates elsewhere gdpr is not the only privacy mandate on earth and then there's on chain applications that you'll word the data what data sets will you store in blockchain you mentioned metadata now metadata I'm not sure because metadata quite often is is updated for lots of reasons for lots of operational patience but really fundamentally if we look at what a blockchain is it's a audit log it's an archive potentially of a just industry fashioned historical data that never changes and you don't want it to change ideally I mean I get an audit log you know let's say in the Internet of Things autonomous vehicles crashed and so forth and the data on how they operate should be stored you know either in a black box on the devices on the cars themself and also possibly backed up to a distributed blockchain where there is a transact or there's a there they a trusted persistent resilient record of what went on that would be a perfect idea for using block chains for storing perhaps trusted timestamp maybe encrypted records on things like that because ultimately the regulators and the courts and the lawyers and everybody else will want to come back and subpoena and use those records to and analyze what went on I mean for example that's an idea where something like a block shape and simile might be employed that doesn't necessarily have to involve PII unless of course it's an individual persons car and so there's all those great areas for those kinds of applications so right now it's kind of looking fuzzy for blockchain in lots of applications where identity can be either you know where you can infer easily the infer the identity of individuals from data that may not on the face of it look like it's PII so Neal I want to come back to you because it's this notion of being able to infer one of the things that's been going on in the industry for the past well 60 years is the dream of being able to create a transaction and persist that data but then generate derivative you out of that data through things like analytics data sharing etc blockchain because it is but you know it basically locks that data away from prying eyes it kind of suggests that we want to be careful about utilizing blockchain for applications where the data could have significant or could generate significant derivative use what do you think well we've known for a long long time that if you have anonymized data in the data set that it can merge that data with data from another data set relatively easy to find out who the individuals are right you add you add DNA stuff to that eh our records surveys things from social media you know everything about people and that's dangerous because we used to think that while losing are losing our privacy means that are going to keep giving us recommendations to buy these hands and shoes it's much more sinister than that you can be discriminated against in employment in insurance in your credit rating and all sorts of things so it's it's I think a really burning issue but what does it have to do with blockchain and G GD R that's an important question I think that blockchain is a really emerge short technology right now and like all image search technologies it's either going to evolve very quickly or it's gonna wither and die I'm not going to speculate which one it's going to be but this issue of how you can use it and how you can monetize data and things that are immutable I think they're all unanswered questions for the wider role of applications but to me it seems like you can get away from the immutable part by taking previous information and simply locking it away with encryption or something else and adding new information the problem becomes I think what happens to that data once someone uses it for other purpose than putting it in a ledger and the other question I have about GD d are in blockchain is who's enforcing this one army of people are sifting through all the stated at the side use and violation does it take a breach before they have it or is there something else going on the act of participating in a blockchain equivalent to owning or or having some visibility or something into a system so I am gdpr again hasn't doesn't seem to have answers to that question Jim what were you gonna say yeah the EU and its member nations have not worked out have not worked out those issues in terms of how will you know they monitor enforcement and enforce GDP are in practical terms I mean clearly it's gonna require on the parts of Germany and France and the others and maybe you know out of Brussels there might be some major Directorate for GDP our monitoring and oversight in terms of you know both companies operating in those nations as well as overseas with European Berger's none of that's been worked out by those nations clearly that's like you know it's just like the implementation issues like blockchain are not blockchain it's we're moving it toward the end of the month with you know not only those issues networked out many companies many enterprises both in Europe and elsewhere are not GDP are ready there may be some of them I'm not gonna name names may make a good boast that they are but know nobody really knows what it needs to be ready at this point I just this came to me very clearly when I asked Bernard Marr well-known author and you know influencer and the big data space at UM in Berlin a few weeks ago at at the data works and I said Bernard you know you consult all over with big companies what percentage of your clients and without giving names do you think are really truly GDP are already perm age when he said very few because they're not sure what it means either everybody's groping their way towards some kind of a hopefully risk mitigations threatened risk mitigation strategy for you know addressing this issue well the technology certainly is moving faster than the law and I'd say an argue even faster than the ethics it's going to be very interesting to see how things play out so we're just for anybody that's interested we are actually in the midst right now of doing right now doing some a nice piece of research on blockchain patterns for applications and what we're talking about essentially here is the idea that blockchain will be applicable to certain classes of applications but a whole bunch of other applications it will not be applicable to so it's another example of a technology that initially people go oh wow that's the technology it's going to solve all problems all date is going to move into the cloud Jim you like to point out Hadoop all data and all applications are going to migrate to the doop and clearly it's not going to happen Neil the way I would answer the question is it blockchain reduces the opportunity for multiple parties to enter into opportunism so that you can use a blockchain as a basis for assuring certain classes of behaviors as a group as a community and and and B and had that be relatively audible and understandable so it can reduce the opportunity for opportunism so you know companies like IBM probably you're right that the idea of a supply chain oriented blockchain that's capable of of assuring that all parties when they are working together are not exploiting holes in the contracts that they're actually complying in getting equal value out of whatever that blockchain system is and they're not gaining it while they can go off and use their own data to do other things if they want that's kind of the in chain and out of chain notion so it's going to be very interesting to see what happens over the course of next few years but clearly even in the example that I described the whole question of gdb our compliance doesn't go away all right so let's get to some action items here Nia what's your action item I suppose but when it comes to gdpr and blockchain I just have a huge number of questions about how they're actually going to be able to enforce it and when it comes to a personal information you know back in the Middle Ages when we went to the market to buy a baby pig they put it in a bag and tied it because they wouldn't want the piglet to run away because it'd take too much trouble to find it but when you got at home sometimes they actually didn't give you a pig they gave you a cat and when you opened up bag the cat was out of the bag that's where the phrase comes from so I'm just waiting for the cat to come out of the bag I I think this sounds like a real fad that was built around Bitcoin and we're trying to find some way to use it in some other way but I'm I just don't know what it is I'm not convinced Jim oxidiser my yeah my advice for Dana managers is to start to segment your data sets into those that are forgettable under gdpr and those that are unforgettable but forgettable ones is anything that has publicly identifiable information or that can be easily aggregated into identifying specific attributes and specific people whether they're in Europe or elsewhere is a secondary issue The Unforgettable is a stuff that it has to remain inviolate and persistent and can that be deleted and so forth the stuff all the unforgettables are suited to writing to one or more locked chains but they are not kosher with gdpr and other privacy mandates and focusing on the unforgettable data whatever that might be then conceivably investigate using blockchain for distributed you know you know access and so forth but they're mine the blockchain just one database technology among many in a very hybrid data architecture you got the Whitman way to skin the cat in terms of HDFS versus blockchain versus you know you know no first no sequel variants don't imagine because blockchain is the flavor of mania of the day that you got to go there there's lots and lots of alternatives all right so here's our action item overall this week we discussed on action item the coming confrontation between gdpr which is has been in effect for a while but actually fines will start being levied after May 25th and blockchain GPR has relatively or prescribed relatively script strict rules regarding a firm's control over personally identifiable in from you have to have it stored within the bounds of the EU if it's derives from an EU source and also it has to be forgettable that source if they choose to be forgotten the firm that owns that data or administers and stewards that data has to be able to get rid of it this is in conflict with blockchain which says that the Ledger's associated with a blockchain will be first of all fully distributed and second of all immutable and that provides some very powerful application opportunities but it's not gdpr compliant on the face of it over the course of the next few years no doubt we will see the EU and other bodies try to bring blockchain and block thing related technologies into a regulatory regime that actually is administrable as as well as auditable and enforceable but it's not there yet does that mean that folks in the EU should not be thinking about blockchains we don't know it means it introduces a risk that has to be accommodated but we at least think that the that what has to happen is data managers on a global basis need to start adding to it this concept of forgettable data and unforgettable data to ensure the cake can remain in compliance the final thing will say is that ultimately blockchain is another one of those technologies that has great science-fiction qualities to it but when you actually start thinking about how you're going to deploy it there are very practical realities associated with what it means to build an application on top of a blockchain datastore ultimately our expectation is that blockchain will be an important technology but it's going to take a number of years for knowledge to diffuse about what blockchain actually is suitable for and what it's not suitable for and this question of gdpr and blockchain interactions is going to be a important catalyst to having some of those conversations once again Neil Jim thank you very much for participating in action today my pleasure I'm Peter burger I'm Peter bursts and you've been once again listening to a wiki bond action item until we talk again

hi I'm Peter Burroughs welcome to a wiki bond action item quick take Neal Raiden Terry data announced earnings this week what does it tell us about Terry data and the overall market for analytics well tear date announced their first quarter earnings and they beat estimates for both earnings than revenues but they but lo they announced lower guidance for the fiscal year which I guess you know failed to impress Wall Street but recurring quarter one revenue was up 11% nearly a year to three hundred and two million dollars but perpetual revenue was down 23% from quarter one seventeen consulting was up to 135 million for the quarter you know not not altogether shabby for a company in transition but I think what it shows is that Teradata is executing this transitional program and there are some pluses and minuses but they're making progress jury's out but I think overall I'd consider it a good quarter what does it tell us about the market anything we can glean from their daters results about the market overall Neal it's hard to say there's a lot of you know at the ATW conference last week I listened to the keynote from Mike Ferguson I've known Mike for years and I think I always think that Mike's the real deal because he spends all of his time doing consulting and when he speaks he's there to tell us what's happening it he gave a great presentation about datawarehouse versus data Lake and if if he's correct there is still a market for a company like Terra data so you know we'll just have to see excellent Neil Raiden thanks very much this has been a wiki bond critique or actually it's been a wiki bond action item quick-take talk to you again

>> Hi, I'm Peter Burris and welcome to another Wikibon Action Item. (upbeat digital music) >> Cameraman: Three, two, one. >> Hi. Once again, we're broadcasting from our beautiful Palo Alto studios, theCUBE studios, and this week we've got another great group. David Floyer in the studio with me along with George Gilbert. And on the phone we've got Jim Kobielus and Ralph Finos. Hey, guys. >> Hi there. >> So we're going to talk about something that's going to become a big issue. It's only now starting to emerge. And that is, what will be the roadmap to automation? Automation is going to be absolutely crucial for the success of IT in the future and the success of any digital business. At its core, many people have presumed that automation was about reducing labor. So introducing software and other technologies, we would effectively be able to substitute for administrative, operator, and related labor. And while that is absolutely a feature of what we're talking about, the bigger issue is ultimately is that we cannot conceive of more complex workloads that are capable of providing better customer experience, superior operations, all the other things a digital business ultimately wants to achieve. If we don't have a capability for simplifying how those underlying resources get put together, configured, or organized, orchestrated, and ultimately sustained delivery of. So the other part of automation is to allow for much more work that can be performed on the same resources much faster. It's a basis for how we think about plasticity and the ability to reconfigure resources very quickly. Now, the challenge is this industry, the IT industry has always used standards as a weapon. We use standards as a basis of creating eco systems or scale, or mass for even something as, like mainframes. Where there weren't hundreds of millions of potential users. But IBM was successful at using that as a basis for driving their costs down and approving a superior product. That's clearly what Microsoft and Intel did many years ago, was achieve that kind of scale through the driving more, and more, and more, ultimately, volume of the technology, and they won. But along the way though, each time, each generation has featured a significant amount of competition at how those interfaces came together and how they worked. And this is going to be the mother of all standard-oriented competition. How does one automation framework and another automation framework fit together? One being able to create value in a way that serves another automation framework, but ultimately as a, for many companies, a way of creating more scale onto their platform. More volume onto that platform. So this notion of how automation is going to evolve is going to be crucially important. David Floyer, are APIs going to be enough to solve this problem? >> No. That's a short answer to that. This is a very complex problem, and I think it's worthwhile spending a minute just on what are the component parts that need to be brought together. We're going to have a multi-cloud environment. Multiple private clouds, multiple public clouds, and they've got to work together in some way. And the automation is about, and you've got the Edge as well. So you've got a huge amount of data all across all of these different areas. And automation and orchestration across that, are as you said, not just about efficiency, they're about making it work. Making it able to be, to work and to be available. So all of the issues of availability, of security, of compliance, all of these difficult issues are a subject to getting this whole environment to be able to work together through a set of APIs, yes, but a lot lot more than that. And in particular, when you think about it, to me, volume of data is critical. Is who has access to that data. >> Peter: Now, why is that? >> Because if you're dealing with AI and you're dealing with any form of automation like this, the more data you have, the better your models are. And if you can increase that amount of data, as Google show every day, you will maintain that handle on all that control over that area. >> So you said something really important, because the implied assumption, and obviously, it's a major feature of what's going on, is that we've been talking about doing more automation for a long time. But what's different this time is the availability of AI and machine learning, for example, >> Right. as a basis for recognizing patterns, taking remedial action or taking predictive action to avoid the need for remedial action. And it's the availability of that data that's going to improve the quality of those models. >> Yes. Now, George, you've done a lot of work around this a whole notion of ML for ITOM. What are the kind of different approaches? If there's two ways that we're looking at it right now, what are the two ways? >> So there are two ends of the extreme. One is I want to see end to end what's going on across my private cloud or clouds. As well as if I have different applications in different public clouds. But that's very difficult. You get end-to-end visibility but you have to relax a lot of assumptions about what's where. >> And that's called the-- >> Breadth first. So the pro is end-to-end visibility. Con is you don't know how all the pieces fit together quite as well, so you get less fidelity in terms of diagnosing root causes. >> So you're trying to optimize at a macro level while recognizing that you can't optimize at a micro level. >> Right. Now the other approach, the other end of the spectrum, is depth first. Where you constrain the set of workloads and services that you're building and that you know about, and how they fit together. And then the models, based on the data you collect there, can become so rich that you have very very high fidelity root cause determination which allows you to do very precise recommendations or even automated remediation. What we haven't figured out hot to do yet is marry the depth first with the breadth first. So that you have multiple focus depth first. That's very tricky. >> Now, if you think about how the industry has evolved, we wrote some stuff about what we call, what I call the iron triangle. Which is basically a very tight relationship between specialists in technology. So the people who were responsible for a particular asset, be it storage, or the system, or the network. The vendors, who provided a lot of the knowledge about how that worked, and therefore made that specialist more or less successful and competent. And then the automation technology that that vendor ultimately provided. Now, that was not automation technology that was associated with AI or anything along those lines. It was kind of out of the box, buy our tool, and this is how you're going to automate various workflows or scripts, or whatever else it might be. And every effort to try to break that has been met with screaming because, well, you're now breaking my automation routines. So the depth-first approach, even without ML, has been the way that we've done it historically. But, David, you're talking about something different. It's the availability of the data that starts to change that. >> Yeah. >> So are we going to start seeing new compacts put in place between users and vendors and OEMs and a lot of these other folks? And it sounds like it's going to be about access to the data. >> Absolutely. So you're going to start. let's start at the bottom. You've got people who have a particular component, whatever that component is. It might be storage. It might be networking. Whatever that component is. They have products in that area which will be collecting data. And they will need for their particular area to provide a degree of automation. A degree of capability. And they need to do two things. They need to do that optimization and also provide data to other people. So they have to have an OEM agreement not just for the equipment that they provide, but for the data that they're going to give and the data they're going to give back. The automatization of the data, for example, going up and the availability of data to help themselves. >> So contracts effectively mean that you're going to have to negotiate value capture on the data side as well as the revenue side. >> Absolutely. >> The ability to do contracting historically has been around individual products. And so we're pretty good at that. So we can say, you will buy this product. I'm delivering you the value. And then the utility of that product is up to you. When we start going to service contracts, we get a little bit different kind of an arrangement. Now, it's an ongoing continuous delivery. But for the most part, a lot of those service contracts have been predicated to known in advance classes of functions, like Salesforce, for example. Or the SASS business where you're able to write a contract that says over time you will have access to this service. When we start talking about some of this automation though, now we're talking about ongoing, but highly bespoke, and potentially highly divergent, over a relatively short period of time, that you have a hard time writing contracts that will prescribe the range of behaviors and the promise about how those behaviors are actually going to perform. I don't think we're there yet. What do you guys think? >> Well, >> No, no way. I mean, >> Especially when you think about realtime. (laughing) >> Yeah. It has to be realtime to get to the end point of automating the actual reply than the actual action that you take. That's where you have to get to. You can't, It won't be sufficient in realtime. I think it's a very interesting area, this contracts area. If you think about solutions for it, I would be going straight towards blockchain type architectures and dynamic blockchain contracts that would have to be put in place. >> Peter: But they're not realtime. >> The contracts aren't realtime. The contracts will never be realtime, but the >> Accessed? access to the data and the understanding of what data is required. Those will be realtime. >> Well, we'll see. I mean, the theorem's what? Every 12 seconds? >> Well. That's >> Everything gets updated? >> That's To me, that's good enough. >> Okay. >> That's realtime enough. It's not going to solve the problem of somebody >> Peter: It's not going to solve the problem at the edge. >> At the very edge, but it's certainly sufficient to solve the problem of contracts. >> Okay. >> But, and I would add to that and say, in addition to having all this data available. Let's go back like 10, 20 years and look at Cisco. A lot of their differentiation and what entrenched them was sort of universal familiarity with their admin interfaces and they might not expose APIs in a way that would make it common across their competitors. But if you had data from them and a constrained number of other providers for around which you would build let's say, these modern big data applications. It's if you constrain the problem, you can get to the depth first. >> Yeah, but Cisco is a great example of it's an archetype for what I said earlier, that notion of an iron triangle. You had Cisco admins >> Yeah. that were certified to run Cisco gear and therefore had a strong incentive to ensure that more Cisco gear was purchased utilizing a Cisco command line interface that did incorporate a fair amount of automation for that Cisco gear and it was almost impossible for a lot of companies to penetrate that tight arrangement between the Cisco admin that was certified, the Cisco gear, and the COI. >> And the exact same thing happened with Oracle. The Oracle admin skillset was pervasive within large >> Peter: Happened with everybody. >> Yes, absolutely >> But, >> Peter: The only reason it didn't happen in the IBM mainframe, David, was because of a >> It did happen, yeah, >> Well, but it did happen, but governments stepped in and said, this violates antitrust. And IBM was forced by law, by court decree, to open up those interfaces. >> Yes. That's true. >> But are we going to see the same type of thing >> I think it's very interesting to see the shape of this market. When we look a little bit ahead. People like Amazon are going to have IAS, they're going to be running applications. They are going to go for the depth way of doing things across, or what which way around is it? >> Peter: The breadth. They're going to be end to end. >> But they will go depth in individual-- >> Components. Or show of, but they will put together their own type of things for their services. >> Right. >> Equally, other players like Dell, for example, have a lot of different products. A lot of different components in a lot of different areas. They have to go piece by piece and put together a consortium of suppliers to them. Storage suppliers, chip suppliers, and put together that outside and it's going to have to be a different type of solution that they put together. HP will have the same issue there. And as of people like CA, for example, who we'll see an opportunity for them to be come in again with great products and overlooking the whole of all of this data coming in. >> Peter: Oh, sure. Absolutely. >> So there's a lot of players who could be in this area. Microsoft, I missed out, of course they will have the two ends that they can combine together. >> Well, they may have an advantage that nobody else has-- >> Exactly. Yeah. because they're strong in both places. But I have Jim Kobielus. Let me check, are you there now? Do we got Jim back? >> Can you hear me? >> Peter: I can barely hear you, Jim. Could we bring Jim's volume up a little bit? So, Jim, I asked the question earlier, about we have the tooling for AI. We know how to get data. How to build models and how to apply the models in a broad brush way. And we're certainly starting to see that happen within the IT operations management world. The ITOM world, but we don't yet know how we're going to write these contracts that are capable of better anticipating, putting in place a regime that really describes how the, what are the limits of data sharing? What are the limits of derivative use? Et cetera. I argued, and here in the studio we generally agreed, that's we still haven't figured that out and that this is going to be one of the places where the tension between, at least in the B2B world, data availability and derivative use and where you capture value and where those profitables go, is going to be significant. But I want to get your take. Has the AI community >> Yeah. started figuring out how we're going to contractually handle obligations around data, data use, data sharing, data derivative use. >> The short answer is, no they have not. The longer answer is, that can you hear me, first of all? >> Peter: Barely. >> Okay. Should I keep talking? >> Yeah. Go ahead. >> Okay. The short answer is, no that the AI community has not addressed those, those IP protection issues. But there is a growing push in the AI community to leverage blockchain for such requirements in terms of block chains to store smart contracts where related to downstream utilization of data and derivative models. But that's extraordinarily early on in its development in terms of insight in the AI community and in the blockchain community as well. In other words, in fact, in one of the posts that I'm working on right now, is looking at a company called 8base that's actually using blockchain to store all of those assets, those artifacts for the development and lifecycle along with the smart contracts to drive those downstream uses. So what I'm saying is that there's lots of smart people like yourselves are thinking about these problems, but there's no consensus, definitely, in the AI community for how to manage all those rights downstream. >> All right. So very quickly, Ralph Finos, if you're there. I want to get your perspective >> Yeah. on what this means from markets, market leadership. What do you think? How's this going to impact who are the leaders, who's likely to continue to grow and gain even more strength? What're your thoughts on this? >> Yeah. I think, my perspective on this thing in the near term is to focus on simplification. And to focus on depth, because you can get return, you can get payback for that kind of work and it simplifies the overall picture so when you're going broad, you've got less of a problem to deal with. To link all these things together. So I'm going to go with the Shaker kind of perspective on the world is to make things simple. And to focus there. And I think the complexity of what we're talking about for breadth is too difficult to handle at this point in time. I don't see it happening any time in the near future. >> Although there are some companies, like Splunk, for example, that are doing a decent job of presenting a more of a breadth approach, but they're not going deep into the various elements. So, George, really quick. Let's talk to you. >> I beg to disagree on that one. >> Peter: Oh! >> They're actually, they built a platform, originally that was breadth first. They built all these, essentially, forwarders which could understand the formats of the output of all sorts of different devices and services. But then they started building what they called curated experiences which is the equivalent of what we call depth first. They're doing it for IT service management. They're doing it for what's called user behavior. Analytics, which is it's a way of tracking bad actors or bad devices on a network. And they're going to be pumping out more of those. What's not clear yet, is how they're going to integrate those so that IT service management understands security and vice versa. >> And I think that's one of the key things, George, is that ultimately, the real question will be or not the real question, but when we think about the roadmap, it's probably that security is going to be early on one of the things that gets addressed here. And again, it's not just security from a perimeter standpoint. Some people are calling it a software-based perimeter. Our perspective is the data's going to go everywhere and ultimately how do you sustain a zero trust world where you know your data is going to be out in the clear so what are you going to do about it? All right. So look. Let's wrap this one up. Jim Kobielus, let's give you the first Action Item. Jim, Action Item. >> Action Item. Wow. Action Item Automation is just to follow the stack of assets that drive automation and figure out your overall sharing architecture for sharing out these assets. I think the core asset will remain orchestration models. I don't think predictive models in AI are a huge piece of the overall automation pie in terms of the logic. So just focus on building out and protecting and sharing and reusing your orchestration models. Those are critically important. In any domain. End to end or in specific automation domains. >> Peter: David Floyer, Action Item. >> So my Action Item is to acknowledge that the world of building your own automation yourself around a whole lot of piece parts that you put together are over. You won't have access to a sufficient data. So enterprises must take a broad view of getting data, of getting components that have data be giving them data. Make contracts with people to give them data, masking or whatever it is and become part of a broader scheme that will allow them to meet the automation requirements of the 21st century. >> Ralph Finos, Action Item. >> Yeah. Again, I would reiterate the importance of keeping it simple. Taking care of the depth questions and moving forward from there. The complexity is enormous, and-- >> Peter: George Gilbert, Action Item. >> I say, start with what customers always start with with a new technology, which is a constrained environment like a pilot and there's two areas that are potentially high return. One is big data, where it's been a multi vendor or multi-vendor component mix, and a mess. And so you take that and you constrain that and make that a depth-first approach in the cloud where there is data to manage that. And the second one is security, where we have now a more and more trained applications just for that. I say, don't start with a platform. Start with those solutions and then start adding more solutions around that. >> All right. Great. So here's our overall Action Item. The question of automation or roadmap to automation is crucial for multiple reasons. But one of the most important ones is it's inconceivable to us to envision how a business can institute even more complex applications if we don't have a way of improving the degree of automation on the underlying infrastructure. How this is going to play out, we're not exactly sure. But we do think that there are a few principals that are going to be important that users have to focus on. Number one is data. Be very clear that there is value in your data, both to you as well as to your suppliers and as you think about writing contracts, don't write contracts that are focused on a product now. Focus on even that product as a service over time where you are sharing data back and forth in addition to getting some return out of whatever assets you've put in place. And make sure that the negotiations specifically acknowledge the value of that data to your suppliers as well. Number two, that there is certainly going to be a scale here. There's certainly going to be a volume question here. And as we think about where a lot of the new approaches to doing these or this notion of automation, is going to come out of the cloud vendors. Once again, the cloud vendors are articulating what the overall model is going to look like. What that cloud experience is going to look like. And it's going to be a challenge to other suppliers who are providing an on-premises true private cloud and Edge orientation where the data must live sometimes it is not something that they just want to do because they want to do it. Because that data requires it to be able to reflect that cloud operating model. And expect, ultimately, that your suppliers also are going to have to have very clear contractual relationships with the cloud players and each other for how that data gets shared. Ultimately, however, we think it's crucially important that any CIO recognized that the existing environment that they have right now is not converged. The existing environment today remains operators, suppliers of technology, and suppliers of automation capabilities and breaking that up is going to be crucial. Not only to achieving automation objectives, but to achieve a converged infrastructure, hyper converged infrastructure, multi-cloud arrangements, including private cloud, true private cloud, and the cloud itself. And this is going to be a management challenge, goes way beyond just products and technology, to actually incorporating how you think about your shopping, organized, how you institutionalize the work that the business requires, and therefore what you identify as a tasks that will be first to be automated. Our expectation, security's going to be early on. Why? Because your CEO and your board of directors are going to demand it. So think about how automation can be improved and enhanced through a security lens, but do so in a way that ensures that over time you can bring new capabilities on with a depth-first approach at least, to the breadth that you need within your shop and within your business, your digital business, to achieve the success and the results that you want. Okay. Once again, I want to thank David Floyer and George Gilbert here in the studio with us. On the phone, Ralph Finos and Jim Kobielus. Couldn't get Neil Raiden in today, sorry Neil. And I am Peter Burris, and this has been an Action Item. Talk to you again soon. (upbeat digital music)

>> Hi, I'm Peter Burris. Welcome to Wikibon's Action Item. (electronic music) Once again, we're broadcasting from our beautiful theCUBE Studios in Palo Alto, California. Here in the studio with me, George Gilbert, and remote, we have Neil Raden, Jim Kobielus, and David Floyer. Welcome, guys! >> Hey. >> Hi, there. >> We've got a really interesting topic today. We're going to be talking about graph databases, which probably just immediately turned off everybody. But we're actually not going to talk so much about it from a technology standpoint. We're really going to spend most of our time talking about it from the standpoint of the business problems that IT and technology are being asked to address, and the degree to which graph databases, in fact, can help us address those problems, and what do we need to do to actually address them. Human beings tend to think in terms of relationships of things to each other. So what the graph community talks about is graphed-shaped problems. And by graph-shaped problem we might mean that someone owns something and someone owns something else, or someone shares an asset, or it could be any number of different things. But we tend to think in terms of things and the relationship that those things have to other things. Now, the relational model has been an extremely successful way of representing data for a lot of different applications over the course of the last 30 years, and it's not likely to go away. But the question is, do these graph-shaped problems actually lend themselves to a new technology that can work with relational technology to accelerate the rate at which we can address new problems, accelerate the performance of those new problems, and ensure the flexibility and plasticity that we need within the application set, so that we can consistently use this as a basis for going out and extending the quality of our applications as we take on even more complex problems in the future. So let's start here. Jim Kobielus, when we think about graph databases, give us a little hint on the technology and where we are today. >> Yeah, well, graph databases have been around for quite a while in various forms, addressing various core-use cases such as social network analysis, recommendation engines, fraud detection, semantic search, and so on. The graph database technology is essentially very closely related to relational, but it's specialized to, when you think about it, Peter, the very heart of a graph-shaped business problem, the entity relationship polygram. And anybody who's studied databases has mastered, at least at a high level, entity relationship diagrams. The more complex these relationships grow among a growing range of entities, the more complex sort of the network structure becomes, in terms of linking them together at a logical level. So graph database technology was developed a while back to be able to support very complex graphs of entities, and relationships, in order to do, a lot of it's analytic. A lot of it's very focused on fast query, they call query traversal, among very large graphs, to find quick answers to questions that might involve who owns which products that they bought at which stores in which cities and are serviced by which support contractors and have which connections or interrelationships with other products they may have bought from us and our partners, so forth and so on. When you have very complex questions of this sort, they lend themselves to graph modeling. And to some degree, to the extent that you need to perform very complex queries of this sort very rapidly, graph databases, and there's a wide range of those on the market, have been optimized for that. But we also have graph abstraction layers over RDBMSes and multi-model databases. You'll find them running in IBM's databases, or Microsoft Cosmos DB, and so forth. You don't need graph-specialized databases in order to do graph queries, in order to manipulate graphs. That's the issue here. When does a specialized graph database serve your needs better than a non-graph-optimized but nonetheless graph-enabling database? That's the core question. >> So, Neil Raden, let's talk a little bit about the classes of business problems that could in fact be served by representing data utilizing a graph model. So these graph-shaped problems, independent of the underlying technology. Let's start there. What kinds of problems can business people start thinking about solving by thinking in terms of graphs of things and relationships amongst things? >> It all comes down to connectedness. That's the basis of a graph database, is how things are connected, either weakly or strongly. And these connected relationships can be very complicated. They can be based on very complex properties. A relational database is not based on, not only is it not based on connectedness, it's not based on connectedness at all. I'd like to say it's based on un-connectedness. And the whole idea in a relational database is that the intelligence about connectedness is buried in the predicate of a query. It's not in the database itself. So I don't know how overlaying graph abstractions on top of a relational database are a good idea. On the other hand, I don't know how stitching a relational database into your existing operation is going to work, either. We're going to have to see. But I can tell you that a major part of data science, machine learning, and AI is going to need to address the issue of causality, not just what's related to each other. And there's a lot of science behind using graphs to get at the causality problem. >> And we've seen, well, let's come back to that. I want to come back to that. But George Gilbert, we've kind of experienced a similar type of thing back in the '90s with the whole concept of object-orientated databases. They were represented as a way of re-conceiving data. The problem was that they had to go from the concept all the way down to the physical thing, and they didn't seem to work. What happened? >> Well it turns out, the big argument was, with object-oriented databases, we can model anything that's so much richer, especially since we're programming with objects. And it turns out, though, that theoretically, especially at that time, you could model anything down at the physical level or even the logical level in a relational database, and so those code bases were able to handle sort of similar, both ends of the use cases, both ends of the spectrum. But now that we have such extreme demands on our data management, rather than look at a whole application or multiple applications even sharing a single relational database, like some of the big enterprise apps, we have workloads within apps like recommendation engines, or a knowledge graph, which explains the relationship between people, places, and things. Or digital twins, or mapping your IT infrastructure and applications, and how they all hold together. You could do that in a relational database, but in a graph database, you can organize it so that you can have really fast analysis of these structures. But, the trade-off is, you're going to be much more restricted in how you can update the stuff. >> Alright, so we think about what happened, then, with some of the object-orientated technology, the original world database, the database was bound to the application, and the developer used the database to tell the application where to go find the data. >> George: Right. >> Relational data allowed us not to tell the applications where to find things, but rather how to find things, and that was persisted, and was very successful for a long time. Object-orientated technologies, in many respects, went back to the idea that the developer had to be very concrete about telling the application where the data was, but we didn't want to do things that way. Now, something's happened, David Floyer. One of the reasons why we had this challenge of representing data in a more abstract way across a lot of different forms without having it also being represented physically, and therefore a lot of different copies and a lot of different representations of the data which broke systems of record and everything else, was that the underlying technology was focused on just persisting data and not necessarily delivering it into these new types of datas, databases, data models, et cetera. But Flash changes that, doesn't it? Can't we imagine a world in which we can have our data in Flash and then, which is a technology that's more focused on delivering data, and then having that data be delivered to a lot of different representations, including things like graph databases, graph models. Is that accurate? >> Absolutely. In a moment I'll take it even further. I think the first point is that when we were designing real-time applications, transactional applications, we were very constrained, indeed, by the amount of data that we could get to. So, as a database administrator, I used to have a rule which you could, database developers could not issue more than 100 database calls. And the reason was that, they could always do more than that, but the applications became very unstable and they became very difficult to maintain. The cost of maintenance went up a lot. The whole area of Flash allows us to do a number of things, and the area of UniGrid enables us to do a number of things very differently. So that we can, for example, share data and have many different views of it. We can use UniGrid to be able to bring far greater amounts of power, compute power, GPUs, et cetera, to bear on specific workloads. I think the most useful thing to think about this is this type of architecture can be used to create systems of intelligence, where you have the traditional relational databases dealing with systems of record, and then you can have the AI systems, graph systems, all the other components there looking at the best way of providing data and making decisions in real time that can be fed back into the systems of record. >> Alright, alright. So let's-- >> George: I want to add to something on this. >> So, Neil, let me come back to you very quickly, sorry, George. Let me come back to Neil. I want to spend, go back to this question of what does a graph-shaped problem look like? Let's kind of run down it. We talked about AI, what about IoT, guys? Is IoT going to help us, is IoT going to drive this notion of looking at the world in terms of graphs more or less? What do you think, Neil? >> I don't know. I hadn't really thought about it, Peter, to tell you the truth. I think that one thing we leave out when we talk about graphs is we talk about, you know, nodes and edges and relationships and so forth, but you can also build a graph with very rich properties. And one thing you can get from a graph query that you can't get from a relational query, unless you write careful predicate, is it can actually do some thinking for you. It can tell you something you don't know. And I think that's important. So, without being too specific about IoT, I have to say that, you know, streaming data and trying to relate it to other data, getting down to, very quickly, what's going on, root-cause analysis, I think graph would be very helpful. >> Great, and, Jim Kobielus, how about you? >> I think, yeah I think that IoT is tailor-made for, or I should say, graph modeling and graph databases are tailor-made for the IoT. Let me explain. I think the IoT, the graph is very much a metadata technology, it's expressing context in a connected universe. Where the IoT is concerned it's all about connectivity, and so graphs, increasingly complex graphs of, say, individuals and the devices and the apps they use and locations and various contexts and so forth, these are increasingly graph-based. They're hierarchical and shifting and changing, and so in order to contextualize and personalize experience in a graph, in an IoT world, I think graph databases will be embedded in the very fabric of these environments. Microsoft has a strategy they announced about a year ago to build more of an intelligent edge around, a distributed graph across all their offerings. So I think graphs will become more important in this era, undoubtedly. >> George, what do you think? Business problems? >> Business problems on IoT. The knowledge graph that holds together digital twin, both of these lend themselves to graph modeling, but to use the object-oriented databases as an example, where object modeling took off was in the applications server, where you had the ability to program, in object-oriented language, and that mapped to a relational database. And that is an option, not the only one, but it's an option for handling graph-model data like a digital twin or IT operations. >> Well that suggests that what we're thinking about here, if we talk about graph as a metadata, and I think, Neil, this partly answers the question that you had about why would anybody want to do this, that we're representing the output of a relational data as a node in a network of data types or data forms so that the data itself may still be relationally structured, but from an application standpoint, the output of that query is, itself, a thing that is then used within the application. >> But to expand on that, if you store it underneath, as fully normalized, in relational language, laid out so that there's no duplicates and things like that, it gives you much faster update performance, but the really complex queries, typical of graph data models, would be very, very slow. So, once we have, say, more in memory technology, or we can manage under the covers the sort of multiple representations of the data-- >> Well that's what Flash is going to allow us to do. >> Okay. >> What David Floyer just talked about. >> George: Okay. >> So we can have a single, persistent, physical storage >> Yeah. >> but it can be represented in a lot of different ways so that we avoid some of the problems that you're starting to raise. If we had to copy the data and have physical, physical copies of the data on disc in a lot of different places then we would run into all kinds of consistency and update. It would probably break the model. We'd probably come back to the notion of a single data store. >> George: (mumbles) >> I want to move on here, guys. One really quick thing, David Floyer, I want to ask you. If there's, you mentioned when you were database administrator and you put restrictions on how many database actions an application or transaction was allowed to generate. When we think about what a business is going to have to do to take advantage of this, are there any particular, like one thing that we need to think about? What's going to change within an IT organization to take advantage of graph database? And we'll do the action items. >> Right. So the key here is the number of database calls can grow by a factor of probably a thousand times what it is now with what we can see is coming as technologies over the next couple of years. >> So let me put that in context, David. That's a single transaction now generating a hundred thousand, >> Correct. >> a hundred thousand database calls. >> Well, access calls to data. >> Right. >> Whatever type of database. And the important thing here is that a lot of that is going to move out, with the discussion of IoT, to where the data is coming in. Because the quicker you can do that, the earlier you can analyze that data, and you talked about IoT with possible different sources coming in, a simple one like traffic lights, for example. The traffic lights are being affected by the traffic lights around them within the city. Those sort of problems are ideal for this sort of graph database. And having all of that data locally and being processed locally in memory very, very close to where those sensors are, is going to be the key to developing solutions in this area. >> So, Neil, I've got one question from you, or one question for you. I'm going to put you on the spot. I just had a thought. And here's the thought. We talk a lot about, in some of the new technologies that could in fact be employed here, whether it be blockchain or even going back to SOA, but when we talk about what a system is going to have the authority to do about the idea of writing contracts that describe very, very discretely, what a system is or is not going to do. I have a feeling those contracts are not going to be written in relational terms. I have a feeling that, like most legal documents, they will be written in what looks more like graph terms. I'm extending that a little bit, but this has rights to do this at this point in time. Is that also, this notion of incorporating more contracts directly to how systems work, to assure that we have the appropriate authorities laid out. What do you think? Is that going to be easier or harder as a consequence of thinking in terms of these graph-shaped models? >> Boy, I don't know. Again, another thing I hadn't really thought about. But I do see some real gaps in thinking. Let me give you an analogy. OLAP databases came on the scene back in the '90s whatever. People in finance departments and whatever they loved OLAP. What they hated was the lack of scalability. And now what we see now is scalability isn't a problem and OLAP solutions are suddenly bursting out all over the place. So I think there's a role for a mental model of how you model your data and how you use it that's different from the relational model. I think the relational model has prominence and has that advantage of, what's it called? Occupancy or something. But I think that the graph is going to show some real capabilities that people are lacking right now. I think some of them are at the very high end, things, like I said, getting to causality. But I think that graph theory itself is so much richer than the simple concept of graphs that's implemented in graph databases today. >> Yeah, I agree with that totally. Okay, let's do the action item round. Jim Kobielus, I want to start with you. Jim, action item. >> Yeah, for data professionals and analytic professionals, focus on what graphs can't do, cannot do, because you hear a lot of hyperbolic, they're not useful for unstructured data or for machine learning in database. They're not as useful for schema on read. What they are useful for is the same core thing that relational is useful for which is schema on write applied to structured data. Number one. Number two, and I'll be quick on this, focus on the core use cases that are already proven out for graph databases. We've already ticked them off here, social network analysis, recommendation engines, influencer analysis, semantic web. There's a rich range of mature use cases for which semantic techniques are suited. And then finally, and I'll be very quick here, bear in mind that relational databases have been supporting graph modeling, graph traversal and so forth, for quite some time, including pretty much all the core mature enterprise databases. If you're using those databases already, and they can perform graph traversals and so forth reasonably well for your intended application, stick with that. No need to investigate the pure play, graph-optimized databases on the market. However, that said, there's plenty of good ones, including AWS is coming out with Neptune. Please explore the other alternatives, but don't feel like you have to go to a graph database first and foremost. >> Alright. David Floyer, action item. >> Action item. You are going to need to move your data center and your applications from the traditional way of thinking about it, of handling things, which is sequential copies going around, usually taking it two or three weeks. You're going to have to move towards a shared data model where the same set of data can have multiple views of it and multiple uses for multiple different types of databases. >> George Gilbert, action item. >> Okay, so when you're looking at, you have a graph-oriented problem, in other words the data is shaped like a graph, question is what type of database do you use? If you have really complex query and analysis use cases, probably best to use a graph database. If you have really complex update requirements, best to use a combination, perhaps of relational and graph or something like multi-model. We can learn from Facebook where, for years, they've built their source of truth for the social graph on a bunch of sharded MySQL databases with some layers on top. That's for analyzing the graph and doing graph searches. I'm sorry, for updating the graph and maintaining it and its integrity. But for reading the graph, they have an entirely different layer for comprehensive queries and manipulating and traversing all those relationships. So, you don't get a free lunch either way. You have to choose your sweet spots and the trade-offs associated with them. >> Alright, Neil Raden, action item. >> Well, first of all, I don't think the graph databases are subject to a lot of hype. I think it's just the opposite. I think they haven't gotten much hype at all. And maybe we're going to see that. But another thing is, a fundamental difference when you're looking at a graph and a graph query, it uses something called open world reasoning. A relational database uses closed world reasoning. I'll give you an example. Country has capital city. Now you have in your graph that China has capital city Beijing, China has capital city Beijing. That doesn't violate the graph. The graph simply understands and intuits that they're different names for the same thing. Now, if you love to write correlated sub-queries for many, many different relationships, I'd say stick to your relational database. I see unique capabilities in a graph that would be difficult to implement in a relational database. >> Alright. Thank you very much, guys. Let's talk about what the action item is for all of us. This week we talked about graph databases. We do believe that they have enormous potential, but we first off have to draw a distinction between graph theory, which is a way of looking at the world and envisioning and conceptualizing solutions to problems, and graph database technology, which has the advantages of being able, for certain classes of data models, to be able to very quickly both write and read data that is based on relationships and hierarchies and network structures that are difficult to represent in a normalized relational database manager. Ultimately, our expectation is that over the next few years, we're going to see an explosion in the class of business problems that lend themselves to a graph-modeling orientation. IoT is an example, very complex analytics systems will be an example, but it is not the only approach or the only way of doing things. But what is interesting, what is especially interesting, is over the last few years, a change in the underlying hardware technology is allowing us to utilize and expand the range of tools that we might use to support these new classes of applications. Specifically, the move to Flash allows us to sustain a single physical copy of data and then have that be represented in a lot of different ways to support a lot of different model forms and a lot of different application types, without undermining the fundamental consistency and integrity of the data itself. So that is going to allow us to utilize new types of technologies in ways that we haven't utilized before, because before, whether it was object-oriented technology or OLAP technology, there was always this problem of having to create new physical copies of data which led to enormous data administrative nightmares. So looking forward, the ability to use Flash as a basis for physically storing the data and delivering it out to a lot of different model and tool forms creates an opportunity for us to use technologies that, in fact, may more naturally map to the way that human beings think about things. Now, where is this likely to really play? We talked about IoT, we talked about other types of technologies. Where it's really likely to play is when the domain expertise of a business person is really pushing the envelope on the nature of the business problem. Historically, applications like accounting or whatnot, were very focused on highly stylized data models, things that didn't necessarily exist in the real world. You don't have double-entry bookkeeping running in the wild. You do have it in the legal code, but for some of the things that we want to build in the future, people, the devices they own, where they are, how they're doing things, that lends itself to a real-world experience and human beings tend to look at those using a graph orientation. And the expectations over the next few years, because of the changes in the physical technology, how we can store data, we will be able to utilize a new set of tools that are going to allow us to more quickly bring up applications, more naturally manage data associated with those applications, and, very important, utilize targeted technology in a broader set of complex application portfolios that are appropriate to solve that particular part of the problem, whether it's a recommendation engine or something else. Alright, so, once again, I want to thank the remote guys, Jim Kobielus, Neil Raden, and David Floyer. Thank you very much for being here. George Gilbert, you're in the studio with me. And, once again, I'm Peter Burris and you've been listening to Action Item. Thank you for joining us and we'll talk to you again soon. (electronic music)

>> Hi I'm Peter Burris and welcome to another Wikibon Action Item. (electronic music) One of the most pressing strategic issues that businesses face is how to get more value out of their data, In our opinion that's the essence of a digital business transformation, is the using of data as an asset to improve your operations and take better advantage of market opportunities. The problem of data though, it's shareable, it's copyable, it's reusable. It's easy to create derivative value out of it. One of the biggest misnomers in the digital business world is the notion that data is the new fuel or the new oil. It's not, You can only use oil once. You can apply it to a purpose and not multiple purposes. Data you can apply to a lot of purposes, which is why you are able to get such interesting and increasing returns to that asset if you use it appropriately. Now, this becomes especially important for technology companies that are attempting to provide digital business technologies or services or other capabilities to their customers. In the consumer world, it started to reach a head. Questions about Facebook's reuse of a person's data through an ad based business model is now starting to lead people to question the degree to which the information asymmetry about what I'm giving and how they're using it is really worth the value that I get out of Facebook, is something that consumers and certainly governments are starting to talk about. it's also one of the bases for GDPR, which is going to start enforcing significant fines in the next month or so. In the B2B world that question is going to become especially acute. Why? Because as we try to add intelligence to the services and the products that we are utilizing within digital business, some of that requires a degree of, or some sort of relationship where some amount of data is passed to improve the models and machine learning and AI that are associated with that intelligence. Now, some companies have come out and said flat out they're not going to reuse a customer's data. IBM being a good example of that. When Ginni Rometty at IBM Think said, we're not going to reuse our customer's data. The question for the panel here is, is that going to be a part of a differentiating value proposition in the marketplace? Are we going to see circumstances in which companies keep products and services low by reusing a client's data and others sustaining their experience and sustaining a trust model say they won't. How is that going to play out in front of customers? So joining me today here in the studio, David Floyer. >> Hi there. >> And on the remote lines we have Neil Raden, Jim Kobielus, George Gilbert, and Ralph Finos. Hey, guys. >> All: Hey. >> All right so... Neil, let me start with you. You've been in the BI world as a user, as a consultant, for many, many number of years. Help us understand the relationship between data, assets, ownership, and strategy. >> Oh, God. Well, I don't know that I've been in the BI world. Anyway, as a consultant when we would do a project for a company, there were very clear lines of what belong to us and what belong to the client. They were paying us generously. They would allow us to come in to their company and do things that they needed and in return we treated them with respect. We wouldn't take their data. We wouldn't take their data models that we built, for example, and sell them to another company. That's just, as far as I'm concerned, that's just theft. So if I'm housing another company's data because I'm a cloud provider or some sort of application provider and I say well, you know, I can use this data too. To me the analogy is, I'm a warehousing company and independently I go into the warehouse and I say, you know, these guys aren't moving their inventory fast enough, I think I'll sell some of it. It just isn't right. >> I think it's a great point. Jim Kobielus. As we think about the role that data, machine learning play, training models, delivering new classes of services, we don't have a clean answer right now. So what's your thought on how this is likely to play out? >> I agree totally with Neil, first of all. If it's somebody else's data, you don't own it, therefore you can't sell and you can't monetize it, clearly. But where you have derivative assets, like machine learning models that are derivative from data, it's the same phenomena, it's the same issue at a higher level. You can build and train, or should, your machine learning models only from data that you have legal access to. You own or you have license and so forth. So as you're building these derivative assets, first and foremost, make sure as you're populating your data lake, to build and to do the training, that you have clear ownership over the data. So with GDPR and so forth, we have to be doubly triply vigilant to make sure that we're not using data that we don't have authorized ownership or access to. That is critically important. And so, I get kind of queasy when I hear some people say we use blockchain to make... the sharing of training data more distributed and federated or whatever. It's like wait a second. That doesn't solve the issues of ownership. That makes it even more problematic. If you get this massive blockchain of data coming from hither and yon, who owns what? How do you know? Do you dare build any models whatsoever from any of that data? That's a huge gray area that nobody's really addressed yet. >> Yeah well, it might mean that the blockchain has been poorly designed. I think that we talked in one of the previous Action Items about the role that blockchain design's going to play. But moving aside from the blockchain, so it seems as though we generally agree that data is owned by somebody typically and that the ownership of it, as Neil said, means that you can't intercept it at some point in time just because it is easily copied and then generate rents on it yourself. David Floyer, what does that mean from a ongoing systems design and development standpoint? How are we going to assure, as Jim said, not only that we know what data is ours but make sure that we have the right protection strategies, in a sense, in place to make sure that the data as it moves, we have some influence and control over it. >> Well, my starting point is that AI and AI infused products are fueled by data. You need that data, and Jim and Neil have already talked about that. In my opinion, the most effective way of improving a company's products, whatever the products are, from manufacturing, agriculture, financial services, is to use AI infused capabilities. That is likely to give you the best return on your money and businesses need to focus on their own products. That's the first place you are trying to protect from anybody coming in. Businesses own that data. They own the data about your products, in use by your customers, use that data to improve your products with AI infused function and use it before your competition eats your lunch. >> But let's build on that. So we're not saying that, for example, if you're a storage system supplier, since that's a relatively easy one. You've got very, very fast SSDs. Very, very fast NVMe over Fabric. Great technology. You can collect data about how that system is working but that doesn't give you rights to then also collect data about how the customer's using the system. >> There is a line which you need to make sure that you are covering. For example, Call Home on a product, any product, whose data is that? You need to make sure that you can use that data. You have some sort of agreement with the customer and that's a win-win because you're using that data to improve the product, prove things about it. But that's very, very clear that you should have a contractual relationship, as Jim and Neil were pointing out. You need the right to use that data. It can't come beyond the hand. But you must get it because if you don't get it, you won't be able to improve your products. >> Now, we're talking here about technology products which have often very concrete and obvious ownership and people who are specifically responsible for administering them. But when we start getting into the IoT domain or in other places where the device is infused with intelligence and it might be collecting data that's not directly associated with its purpose, just by virtue of the nature of sensors that are out there and the whole concept of digital twin introduces some tension in all this. George Gilbert. Take us through what's been happening with the overall suppliers of technology that are related to digital twin building, designing, etc. How are they securing or making promises committing to their customers that they will not cross this data boundary as they improve the quality of their twins? >> Well, as you quoted Ginni Rometty starting out, she's saying IBM, unlike its competitors, will not take advantage and leverage and monetize your data. But it's a little more subtle than that and digital twins are just sort of another manifestation of industry-specific sort of solution development that we've done for decades. The differences, as Jim and David have pointed out, that with machine learning, it's not so much code that's at the heart of these digital twins, it's the machine learning models and the data is what informs those models. Now... So you don't want all your secret sauce to go from Mercedes Benz to BMW but at the same time the economics of industry solutions means that you do want some of the repeatability that we've always gotten from industry solutions. You might have parts that are just company specific. And so in IBM's case, if you really parse what they're saying, they take what they learn in terms of the models from the data when they're working with BMW, and some of that is going to go into the industry specific models that they're going to use when they're working with Mercedes-Benz. If you really, really sort of peel the onion back and ask them, it's not the models, it's not the features of the models, but it's the coefficients that weight the features or variables in the models that they will keep segregated by customer. So in other words, you get some of the benefits, the economic benefits of reuse across customers with similar expertise but you don't actually get all of the secret sauce. >> Now, Ralph Finos-- >> And I agree with George here. I think that's an interesting topic. That's one of the important points. It's not kosher to monetize data that you don't own but conceivably if you can abstract from that data at some higher level, like George's describing, in terms of weights and coefficients and so forth, in a neural network that's derivative from the model. At some point in the abstraction, you should be able to monetize. I mean, it's like a paraphrase of some copyrighted material. A paraphrase, I'm not a lawyer, but you can, you can sell a paraphrase because it's your own original work that's based obviously on your reading of Moby Dick or whatever it is you're paraphrasing. >> Yeah, I think-- >> Jim I-- >> Peter: Go ahead, Neil. >> I agree with that but there's a line. There was a guy who worked at Capital One, this was about ten years ago, and he was their chief statistician or whatever. This was before we had words like machine learning and data science, it was called statistics and predictive analytics. He left the company and formed his own company and rewrote and recoded all of the algorithms he had for about 20 different predictive models. Formed a company and then licensed that stuff to Sybase and Teradata and whatnot. Now, the question I have is, did that cross the line or didn't it? These were algorithms actually developed inside Capital One. Did he have the right to use those, even if he wrote new computer code to make them run in databases? So it's more than just data, I think. It's a, well, it's a marketplace and I think that if you own something someone should not be able to take it and make money on it. But that doesn't mean you can't make an agreement with them to do that, and I think we're going to see a lot of that. IMSN gets data on prescription drugs and IRI and Nielsen gets scanner data and they pay for it and then they add value to it and they resell it. So I think that's really the issue is the use has to be understood by all the parties and the compensation has to be appropriate to the use. >> All right, so Ralph Finos. As a guy who looks at market models and handles a lot of the fundamentals for how we do our forecasting, look at this from the standpoint of how people are going to make money because clearly what we're talking about sounds like is the idea that any derivative use is embedded in algorithms. Seeing how those contracts get set up and I got a comment on that in a second, but the promise, a number of years ago, is that people are going to start selling data willy-nilly as a basis for their economic, a way of capturing value out of their economic activities or their business activities, hasn't matured yet generally. Do we see like this brand new data economy, where everybody's selling data to each other, being the way that this all plays out? >> Yeah, I'm having a hard time imagining this as a marketplace. I think we pointed at the manufacturing industries, technology industries, where some of this makes some sense. But I think from a practitioner perspective, you're looking for variables that are meaningful that are in a form you can actually use to make prediction. That you understand what the the history and the validity of that of that data is. And in a lot of cases there's a lot of garbage out there that you can't use. And the notion of paying for something that ultimately you look at and say, oh crap, it's not, this isn't really helping me, is going to be... maybe not an insurmountable barrier but it's going to create some obstacles in the market for adoption of this kind of thought process. We have to think about the utility of the data that feeds your models. >> Yeah, I think there's going to be a lot, like there's going to be a lot of legal questions raised and I recommend that people go look at a recent SiliconANGLE article written by Mike Wheatley and edited by our Editor In Chief Robert Hof about Microsoft letting technology partners own right to joint innovations. This is a quite a difference. This is quite a change for Microsoft who used to send you, if you sent an email with an idea to them, you'd often get an email back saying oh, just to let you know any correspondence we have here is the property of Microsoft. So there clearly is tension in the model about how we're going to utilize data and enable derivative use and how we're going to share, how we're going to appropriate value and share in the returns of that. I think this is going to be an absolutely central feature of business models, certainly in the digital business world for quite some time. The last thing I'll note and then I'll get to the Action Items, the last thing I'll mention here is that one of the biggest challenges in whenever we start talking about how we set up businesses and institutionalize the work that's done, is to look at the nature of the assets and the scope of the assets and in circumstances where the asset is used by two parties and it's generating a high degree of value, as measured by the transactions against those assets, there's always going to be a tendency for one party to try to take ownership of it. One party that's able to generate greater returns than the other, almost always makes move to try to take more control out of that asset and that's the basis of governance. And so everybody talks about data governance as though it's like something that you worry about with your backup and restore. Well, that's important but this notion of data governance increasingly is going to become a feature of strategy and boardroom conversations about what it really means to create data assets, sustain those data assets, get value out of them, and how we determine whether or not the right balance is being struck between the value that we're getting out of our data and third parties are getting out of our data, including customers. So with that, let's do a quick Action Item. David Floyer, I'm looking at you. Why don't we start here. David Floyer, Action Item. >> So my Action Item is for businesses, you should focus. Focus on data about your products in use by your customers, to improve, help improve the quality of your products and fuse AI into those products as one of the most efficient ways of adding value to it. And do that before your competition has a chance to come in and get data that will stop you from doing that. >> George Gilbert, Action Item. >> I guess mine would be that... in most cases you you want to embrace some amount of reuse because of the economics involved from your joint development with a solution provider. But if others are going to get some benefit from sort of reusing some of the intellectual property that informs models that you build, make sure you negotiate with your vendor that any upgrades to those models, whether they're digital twins or in other forms, that there's a canonical version that can come back and be an upgraded path for you as well. >> Jim Kobielus, Action Item. >> My Action Item is for businesses to regard your data as a product that you monetize yourself. Or if you are unable to monetize it yourself, if there is a partner, like a supplier or a customer who can monetize that data, then negotiate the terms of that monetization in your your relationship and be vigilant on that so you get a piece of that stream. Even if the bulk of the work is done by your partner. >> Neil Raden, Action Item. >> It's all based on transparency. Your data is your data. No one else can take it without your consent. That doesn't mean that you can't get involved in relationships where there's an agreement to do that. But the problem is most agreements, especially when you look at a business consumer, are so onerous that nobody reads them and nobody understands them. So the person providing the data has to have an unequivocal right to sell it to you and the person buying it has to really understand what the limits are that they can do with it. >> Ralph Finos, Action Item. You're muted Ralph. But it was brilliant, whatever it was. >> Well it was and I really can't say much more than that. (Peter laughs) But I think from a practitioner perspective and I understand that from a manufacturing perspective how the value could be there. But as a practitioner if you're fishing for data out there that someone has that might look like something you can use, chances are it's not. And you need to be real careful about spending money to get data that you're not really clear is going to help you. >> Great. All right, thanks very much team. So here's our Action Item conclusion for today. The whole concept of digital business is predicated in the idea of using data assets in a differential way to better serve your markets and improve your operations. It's your data. Increasingly, that is going to be the base for differentiation. And any weak undertaking to allow that data to get out has the potential that someone else can, through their data science and their capabilities, re-engineer much of what you regard as your differentiation. We've had conversations with leading data scientists who say that if someone were to sell customer data into a open marketplace, that it would take about four days for a great data scientist to re-engineer almost everything about your customer base. So as a consequence, we have to tread lightly here as we think about what it means to release data into the wild. Ultimately, the challenge there for any business will be: how do I establish the appropriate governance and protections, not just looking at the technology but rather looking at the overall notion of the data assets. If you don't understand how to monetize your data and nonetheless enter into a partnership with somebody else, by definition that partner is going to generate greater value out of your data than you are. There's significant information asymmetries here. So it's something that, every company must undertake an understanding of how to generate value out of their data. We don't think that there's going to be a general-purpose marketplace for sharing data in a lot of ways. This is going to be a heavily contracted arrangement but it doesn't mean that we should not take great steps or important steps right now to start doing a better job of instrumenting our products and services so that we can start collecting data about our products and services because the path forward is going to demonstrate that we're going to be able to improve, dramatically improve the quality of the goods and services we sell by reducing the assets specificities for our customers by making them more intelligent and more programmable. Finally, is this going to be a feature of a differentiated business relationship through trust? We're open to that. Personally, I'll speak for myself, I think it will. I think that there is going to be an important element, ultimately, of being able to demonstrate to a customer base, to a marketplace, that you take privacy, data ownership, and intellectual property control of data assets seriously and that you are very, very specific, very transparent, in how you're going to use those in derivative business transactions. All right. So once again, David Floyer, thank you very much here in the studio. On the phone: Neil Raden, Ralph Finos, Jim Kobielus, and George Gilbert. This has been another Wikibon Action Item. (electronic music)

>> Hi, I'm Peter Burris with another Wikibon Action Item Quick Take. David Floyer, you've been at the vanguard of talking about the role that Flash, SSD's, and others, other technologies are going to have in the technology industry, predicting early on that it was going to eclipse HDD, even though you got a lot of blow back about the "We're going to remain expensive and small". That's changed. What's going on? >> Well, I've got a prediction that we'll have petabyte drives, SSD drives, within five years. Let me tell you a little bit why. So there's this new type of SSD that's coming into town. It's the mega SSD, and Nimbus Data has just announced this mega SSD. It's a hundred terabyte drive. It's very high density, obviously. It has much fewer, uh, much fewer? It has fewer IOPS and bandwidth than SSD. The access density is much better than HDD, but still obviously lower than high-performance SSD. Much, much lower space power than either SSD or HDD in terms of environmentals. It's three and a half inch. That's compatible with HDD. It's obviously looking to go into the same slots. A hundred terabytes today, two hundred terabytes, 10x, that 10x of the Hammer drives that are coming in from HDD's in 2019, 2020, and the delta will increase over time. It's still more expensive than HDD per bit, but it's, and it's not a direct replacement, but much greater ability to integrate with data services and other things like that. So the prediction, then, is get ready for mega SSD's. It's going to carve out a space at the low end of SSD's and into the HDD's, and we're going to have one petabyte, or more, drives within five years. >> Big stuff from small things. David Floyer, thank you very much. And, once again, this has been a Wikibon Action Item Quick Take. (chill techno music)

>> Hi, I'm Peter Burris and welcome to another Wikibon Action Item. (electronic music) Once again, we're broadcasting from theCUBE studios in beautiful Palo Alto. Here in the studio with me are George Gilbert and David Floyer. And remote, we have Neil Raden and Jim Kobielus. Welcome everybody. >> David: Thank you. >> So this is kind of an interesting topic that we're going to talk about this week. And it really is how are we going to find new ways to generate derivative use out of many of the applications, especially web-based applications that are have been built over the last 20 years. A basic premise of digital business is that the difference between business and digital business is the data and how you craft data as an asset. Well, as we all know in any universal Turing machine, data is the basis for representing both the things that you're acting upon but also the algorithms, the software itself. Software is data and the basic principles of how we capture software oriented data assets or software assets and then turn them into derivative sources of value and then reapply them to new types of problems is going to become an increasingly important issue as we think about the world of digital business is going to play over the course of the next few years. Now, there are a lot of different domains where this might work but one in particular that's especially as important is in the web application world where we've had a lot of application developers and a lot of tools be a little bit more focused on how we use web based services to manipulate things and get software to do the things we want to do and also it's a source of a lot of the data that's been streaming into big data applications. And so it's a natural place to think about how we're going to be able to create derivative use or derivative value out of crucial software assets. How are we going to capture those assets, turn them into something that has a different role for the business, performs different types of work, and then reapply them. So to start the conversation, Jim Kobielus. Why don't you take us through what some of these tools start to look like. >> Hello, Peter. Yes, so really what we're looking at here, in order to capture these assets, the web applications, we first have to generate those applications and the bulk of that worker course is and remains manual. And in fact, there is a proliferation of web application development frameworks on the market and the range of them continues to grow. Everything from React to Angular to Ember and Node.js and so forth. So one of the core issues that we're seeing out there in the development world is... are there too many of these. Is there any prospect for simplification and consolidation and convergence on web application development framework to make the front-end choices for developers a bit easier and straightforward in terms of the front-end development of JavaScript and HTML as well as the back-end development of the logic to handle the interactions; not only with the front-end on the UI side but also with the infrastructure web services and so forth. Once you've developed the applications, you, a professional programmer, then and only then can we consider the derivative uses you're describing such as incorporation or orchestration of web apps through robotic process automation and so forth. So the issue is how can we simplify or is there a trend toward simplification or will there soon be a trend towards simplification of a front-end manual development. And right now, I'm not seeing a whole lot of action in this direction of a simplification on the front-end development. It's just a fact. >> So we're not seeing a lot of simplification and convergence on the actual frameworks for creating software or creating these types of applications. But we're starting to see some interesting trends for stuff that's already been created. How can we generate derivative use out of it? And also per some of our augmented programming research, new ways of envisioning the role that artificial intelligence machine learning, etc, can play in identifying patterns of utilization so that we are better able to target those types of things that could be used for derivative or could be applied to derivative use. Have I got that right, Jim? >> Yeah, exactly. AI within robotic process automation, anything that could has already been built can be captured through natural language processing, through a computer image recognition, OCR, and so forth. And then trans, in that way, it's an asset that can be repurposed in countless ways and that's the beauty RPA or where it's going. So the issue is then not so much capture of existing assets but how can we speed up and really automate the original development of all that UI logic? I think RPA is part of the solution but not the entire solution, meaning RPA provides visual front-end tools for the rest of us to orchestrate more of the front-end development of the application UI and interaction logic. >> And it's also popping up-- >> That's part of broader low-code-- >> Yeah, it's also popping up at a lot of the interviews that we're doing with CIOs about related types of things but I want to scope this appropriately. So we're not talking about how we're going to take those transaction processing applications, David Floyer, and envelope them and containerize them and segment them and apply a new software. That's not what we're talking about, nor are we talking about the machine to machine world. Robot process automation really is a tool for creating robots out of human time interfaces that can scale the amount of work and recombine it in different ways. But we're not really talking about the two extremes. The hardcore IoT or the hardcore systems of record. Right? >> Absolutely. But one question I have for Jim and yourself is the philosophy for most people developing these days is mobile first. The days of having an HTML layout on a screen have gone. If you aren't mobile first, that's going to be pretty well a disaster for any particular development. So Jim, how does RPA and how does your discussion fit in with mobile and all of the complexity that mobile brings? All of the alternative ways that you can do things with mobile. >> Yeah. Well David, of course, low-code tools, there are many. There are dozens out there. There are many of those that are geared towards primarily supporting of fast automated development of mobile applications to run on a variety of devices and you know, mobile UIs. That's part of the solution as it were but also in the standard web application development world. know there's these frameworks that I've described. Everything from React to Angular to Vue to Ember, everything else, are moving towards a concept, more than concept, it's a framework or paradigm called progressive web apps. And what progressive web apps are all about, that's really the mainstream of web application development now is blurring the distinction between mobile and web and desktop applications because you build applications, JavaScript applications for browsers. The apps look and behave as if they were real-time interactive in memory mobile apps. What that means is that they download fresh content throughout a browsing session progressively. I'm putting to determine air quotes because that's where the progressive web app comes in. And they don't require the end-user to visit an app store or download software. They don't require anything in terms of any special capabilities in terms of synchronizing data from servers to run in memory natively inside of web accessible containers that are local to the browser. They just feel mobile even though they, excuse me, they may be running on a standard desktop with narrowband connectivity and so forth. So they scream and they scream in the context of their standard JavaScript Ajax browser obsession. >> So when we think about this it got, jeez Jim it almost sounds like like client-side Java but I think you're we're talking about something, as you said, that that evolves as the customer uses it and there's a lot of techniques and approaches that we've been using to do some of those things. But George Gilbert, the reason I bring up the notion of client-side Java is because we've seen other initiatives over the years try to do this. Now, partly they failed because, David Floyer, they focused on too much and tried to standardize or presume that everything required a common approach and we know that that's always going to fail. But what are some of the other things that we need to think about as we think about ways of creating derivative use out of software or in digital assets. >> Okay, so. I come at it from two angles. And as Jim pointed out, there's been a Cambrian explosion of creativity and innovation on frankly on client-side development and server-side development. But if you look at how we're going to recombine our application assets, we tried 20 years ago with EAI but that was, and it's sort of like MuleSoft but only was for on-prem apps. And it didn't work because every app was bespoke essentially-- >> Well, it worked for point-to-point classes of applications. >> Yeah, but it required bespoke development for every-- >> Peter: Correct. >> Every instance because the apps were so customized. >> Peter: And the interfaces were so customized. >> Yes. At the same time we were trying to build higher-level application development capabilities on desktop productivity tools with macros and then scripting languages, cross application, and visual development or using applications as visual development building blocks. Now, you put those two things together and you have the ability to work with user interfaces by building on, I'm sorry, to work with applications that have user interfaces and you have the functionality that's in the richer enterprise applications and now we have the technology to say let's program by example on essentially a concrete use case and a concrete workflow. And then you go back in and you progressively generalize it so it can handle more exception conditions and edge conditions. In other words, you start with... it's like you start with the concrete and you get progressively more abstract. >> Peter: You start with the work that the application performs. >> Yeah. >> And not knowledge of the application itself. >> Yes. But the key thing is, as you said, recombining assets because we're sort of marrying the best of EAI world with the best of the visual client-side development world. Where, as Jim points out, machine learning is making it easier for the tools to stay up to date as the user interfaces change across releases. This means that, I wouldn't say this as easy as spreadsheet development, it's just not. >> It's not like building spreadsheet macros but it's more along those lines. >> Yeah, but it's not as low-level as just building raw JavaScript because, and Jim's great example of JavaScript client-side frameworks. Look at our Gmail inbox application that millions of people use. That just downloads a new version whenever they want to drop it and they're just shipping JavaScript over to us. But the the key thing and this is, Peter, your point about digital business. By combining user interfaces, we can bridge applications that were silos then we can automate the work the humans were doing to bridge those silos and then we can reconstitute workflows in much more efficient-- >> Around the digital assets, which is kind of how business ultimately evolves. And that's a crucial element of this whole thing. So let's change direction a little bit because we're talking about, as Jim said, we've been talking about the fact that there are all these frameworks out there. There may be some consolidation on the horizon, we're researching that right now. Although there's not a lot of evidence that it's happening but there clearly is an enormous number of digital assets that are in place inside these web-based applications, whether it be relative to mobile or something else. And we want to find derivative use of or we want to create derivative use out of them and there's some new tools that allow us to do that in a relatively simple straightforward way, like RPA and there are certainly others. But that's not where this ends up. We know that this is increasingly going to be a target for AI, what we've been calling augmented programming and the ability to use machine learning and related types of technologies to be able to reveal, make transparent, gain visibility into, patterns within applications and within the use of data and then have that become a crucial feature of the development process. And increasingly even potentially to start actually creating code automatically based on very clear guidance about what work needs to be performed. Jim, what's happening in that world right now? >> Oh, let's see. So basically, I think what's going to happen over time is that more of the development cycle for web applications will incorporate not just the derivative assets, the AI to be able to decompose existing UI elements and recombine them. Enable flexible and automated recombination in various ways but also will enable greater tuning of the UI in an automated fashion through A/B testing that's in line to the development cycle based on metrics that AI is able to sift through in terms of... different UI designs can be put out into production applications in real time and then really tested with different categories of users and then the best suited or best fit a design based on like reducing user abandonment rates and speeding up access to commonly required capabilities and so forth. The metrics can be rolled in line to the automation process to automatically select the best fit UI design that had been developed through automated means. In other words, this real-world experimentation of the UI has been going on for quite some time in many enterprises and it's often, increasingly it involves data scientists who are managing the predictive models to sort of very much drive the whole promotion process of promoting the best fit design to production status. I think this will accelerate. We'll take more of these in line metrics on UI and then we brought, I believe, into more RPA style environments so the rest of us building out these front ends are automating more of our transactions and many more of the UIs can't take advantage of the fact that we'll let the infrastructure choose the best fit of the designs for us without us having to worry about doing A/B testing and all that stuff. The cloud will handle it. >> So it's a big vision. This notion of it, even eventually through more concrete standard, well understood processes to apply some of these AIML technologies to being able to choose options for the developer and even automate some elements of those options based on policy and rules. Neil Raden, again, we've been looking at similar types of things for years. How's that worked in the past and let's talk a bit about what needs to happen now to make sure that if it's going to work, it's going to work this time. >> Well, it really hasn't worked very well. And the reason it hasn't worked very well is because no one has figured out a representational framework to really capture all the important information about these objects. It's just too hard to find them. Everybody knows that when you develop software, 80% of it is grunt work. It's just junk. You know, it's taking out the trash and it's setting things up and whatever. And the real creative stuff is a very small part of it. So if you could alleviate the developer from having to do all that junk by just picking up pieces of code that have already been written and tested, that would be big. But the idea of this has been overwhelmed by the scale and the complexity. And people have tried to create libraries like JavaBeans and object-oriented programming and that sort of thing. They've tried to create catalogs of these things. They've used relational databases, doesn't work. My feeling and I hate to use the word because it always puts people to sleep is some kind of ontology that's deep enough and rich enough to really do this. >> Oh, hold on Neil, I'm feeling... (laughs) >> Yeah. Well, I mean, what good is it, I mean go to Git, right. You can find a thousand things but you don't know which one is really going to work for you because it's not rich enough, it doesn't have enough information. It needs to have quality metrics. It needs to have reviews by people who have used converging and whatever. So that's that's where I think we run into trouble. >> Yeah, I know. >> As far as robots, yeah? >> Go ahead. >> As far as robots writing code, you're going to have the same problem. >> No, well here's where I think it's different this time and I want to throw it out to you guys and see if it's accurate and we'll get to the action items. Here's where I think it's different. In the past, partly perhaps because it's where developers were most fascinated, we try to create object-oriented database and object oriented representations of data and object oriented, using object oriented models as a way of thinking about it. And object oriented code and object oriented this and and a lot of it was relatively low in the stack. And we try to create everything from scratch and it turned out that whenever we did that, it was almost like CASE from many years ago. You create it in the tool and then you maintain it out of the tool and you lose all organization of how it worked. What we're talking about here, and the reason why I think this is different, I think Neil is absolutely right. It's because we're focusing our attention on the assets within an application that create the actual business value. What does the application do and try to encapsulate those and render those as things that are reusable without necessarily doing an enormous amount of work on the back-end. Now, we have to be worried about the back-end. It's not going to do any good to do a whole bunch of RPA or related types of stuff on the front-end that kicks off an enormous number of transactions that goes after a little server that's 15 years old. That's historically only handled a few transactions a minute. So we have to be very careful about how we do this. But nonetheless, by focusing more attention on what is generating value in the business, namely the actions that the application delivers as opposed to knowledge of the application itself, namely how it does it then I think that we're constraining the problem pretty dramatically subject to the realities of what it means to actually be able to maintain and scale applications that may be asked to do more work. What do you guys think about that? >> Now Peter, let me say one more thing about this, about robots. I think you're all a lot more sanguine about AI and robots doing these kinds of things. I'm not. Let me read to you have three pickup lines that a deep neural network developed after being trained to do pickup lines. You must be a tringle? 'Cause you're the only thing here. Hey baby, you're to be a key? Because I can bear your toot? Now, what kind of code would-- >> Well look, the problems look, we go back 50 years and ELIZA and the whole notion of whatever it was. The interactive psychology. Look, let's be honest about this. Neil, you're making a great point. I don't know that any of us are more or less sanguine and that probably is a good topic for a future action item. What are the practical limits of AI and how that's going to change over time. But let's be relatively simple here. The good news about applying AI inside IT problems is that you're starting with engineered systems, with engineered data forms, and engineered data types, and you're working with engineers, and a lot of that stuff is relatively well structured. Certainly more structured than the outside world and it starts with digital assets. That's why a AI for IT operations management is more likely. That's why AI for application programming is more likely to work as opposed to AI to do pickup lines, which is as you said semantically it's all over the place. There's very, very few people that are going to conform to a set of conventions for... Well, I want to move away from the concept of pickup lines and set conventions for other social interactions that are very, very complex. We don't look at a face and get excited or not in a way that corresponds to an obvious well-understood semantic problem. >> Exactly, the value that these applications deliver is in their engagement with the real world of experience and that's not the, you can't encode the real world of human lived experience in a crisp clear way. It simply has to be proven out in the applications or engagement through people or not through people, with the real world outcome and then some outcomes like the ones that Neil read off there, in terms of those ridiculous pickup lines. Most of those kinds of automated solutions won't make a freaking bit of sense because you need humans with their brains. >> Yeah, you need human engagement. So coming back to this key point, the constraint that we're putting on this right now and the reason why certainly, perhaps I'm a little bit more ebullient than you might be Neil. But I want to be careful about this because I also have some pretty strong feelings about where what the limits of AI are, regardless of what Elon Musk says. That at the end of the day, we're talking about digital objects, not real objects, that are engineered, not, haven't evolved over a few billion years, to deliver certain outputs and data that's been tested and relatively well verified. As opposed to have an unlimited, at least from human experience standpoint, potential set of outcomes. So in that small world and certainly the infrastructure universe is part of that and what we're saying is increasingly the application development universe is going to be part of that as part of the digital business transformation. I think it's fair to say that we're going to start seeing AI machine learning and some of these other things being applied to that realm with some degree of success. But, something to watch for. All right, so let's do action item. David Floyer, why don't we start with you. Action item. >> In addressing this, I think that the keys in terms of business focus is first of all mobiles, you have to design things for mobile. So any use of any particular platform or particular set of tools has to lead to mobile being first. And the mobiles are changing rapidly with the amount of data that's being generated on the mobile itself, around the mobile. So that's the first point I would make from a business perspective. And the second is that from a business perspective, one of the key things is that you can reduce cost. Automation must be a key element of this and therefore designing things that will take out tasks and remove tasks, make things more efficient, is going to be an incredibly important part of this. >> And reduce errors. >> And reduce errors, absolutely. Probably most important is reduce errors. Is to take those out of the of the chain and where you can speed things up by removing human intervention and human tasks and raising what humans are doing to a higher level. >> Other things. George Gilbert, action item. >> Okay, so. Really quickly on David's point that we have many more application forms and expressions that we have to present like mobile first. And going back to using RPA as an example. The UiPath product that we've been working with, the core of its capability is to be able to identify specific UI elements in a very complex presentation, whether it's on a web browser or whether it's on a native app on your desktop or whether it's mobile. I don't know how complete they are on mobile because I'm not sure if they did that first but that core capability to identify in a complex, essentially collection and hierarchy of UI elements, that's what makes it powerful. Now on the AI part, I don't think it's as easy as pointing it at one app and then another and say go make them talk. It's more like helping you on the parts where they might be a little ambiguous, like if pieces move around from release to release, things like that. So my action item is say start prototyping with the RPA tools because that's probably, they're probably robust enough to start integrating your enterprise apps. And the only big new wrinkle that's come out in the last several weeks that is now in everyone's consciousness is the MuleSoft acquisition by Salesforce because that's going back to the EAI model. And we will see more app to app integration at the cloud level that's now possible. >> Neil Raden, action item. >> Well, you know, Mark Twain said, there's only two kinds of people in the world. The kind who think there are only two kinds of people in the world and the ones who know better. I'm going to deviate from that a little and say that there's really two kinds of software developers in the world. They're the true computer scientists who want to write great code. It's elegant, it's maintainable, it adheres to all the rules, it's creative. And then there's an army of people who are just trying to get something done. So the boss comes to you and says we've got to get a new website up apologizing for selling the data of 50 million of our customers and you need to do it in three days. Now, those are the kind of people who need access to things that can be reused. And I think there's a huge market for that, as well as all these other software development robots so to speak. >> Jim Kobielus, action item. >> Yeah, for simplifying web application development, I think that developers need to distinguish between back-end and front-end framework. There's a lot of convergence around the back-end framework. Specifically Node.js. So you can basically decouple the decision in terms of front-end frameworks from that and you need to write upfront. Make sure that you have a back-end that supports many front ends because there are many front ends in the world. Secondly, the front ends themselves seem to be moving towards React and Angular and Vue as being the predominant ones. You'll find more programmers who are familiar with those. And then thirdly, as you move towards consolidation on to fewer frameworks on the front-end, move towards low-code tools that allow you just with the push of a button, you know visual development, being able to deploy the built out UI to a full range of mobile devices and web applications. And to close my action item... I'll second what David said. Move toward a mobile first development approach for web applications with a focus on progressive web applications that can run on mobiles and others. Where they give a mobile experience. With intermittent connectivity, with push notifications, with a real-time in memory fast experience. Move towards a mobile first development paradigm for all of your your browser facing applications and that really is the simplification strategy you can and should pursue right now on the development side because web apps are so important, you need a strategy. >> Yeah, so mobile irrespective of the... irrespective of the underlying biology or what have you of the user. All right, so here's our action item. Our view on digital business is that a digital business uses data differently than a normal business. And a digital business transformation ultimately is about how do we increase our visibility into our data assets and find new ways of creating new types of value so that we can better compete in markets. Now, that includes data but it also includes application elements, which also are data. And we think increasingly enterprises must take a more planful and purposeful approach to identifying new ways of deriving additional streams of value out of application assets, especially web application assets. Now, this is a dream that's been put forward for a number of years and sometimes it's work better than others. But in today's world we see a number of technologies emerging that are likely, at least in this more constrained world, to present a significant new set of avenues for creating new types of digital value. Specifically tools like RPA, remote process automation, that are looking at the outcomes of an application and allow programmers use a by example approach to start identifying what are the UI elements, what those UI elements do, how they could be combined, so that they can be composed into new things and thereby provide a new application approach, a new application integration approach which is not at the data and not at the code but more at the work that a human being would naturally do. These allow for greater scale and greater automation and a number of other benefits. The reality though is that you also have to be very cognizant as you do this, even though you can find these, find these assets, find a new derivative form and apply them very quickly to new potential business opportunities that you have to know what's happening at the back-end as well. Whether it's how you go about creating the assets, with some of the front-end tooling, and being very cognizant of which front ends are going to be better or not better or better able at creating these more reusable assets. Or whether you're talking about still how relatively mundane things like how a database serialized has access to data and will fall over because you've created an automated front-end that's just throwing a lot of transactions at it. The reality is there's always going to be complexity. We're not going to see all the problems being solved but some of the new tools allow us to focus more attention on where the real business value is created by apps, find ways to reuse that, and apply it, and bring it into a digital business transformation approach. All right. Once again. George Gilbert, David Floyer, here in the studio. Neil Raden, Jim Kobielus, remote. You've been watching Wikibon Action Item. Until next time, thanks for joining us. (electronic music)

>> Hi, this is Peter Burris with another Wikibon Action Item Quick Take. David Floyer, big news from Redmond, what's going on? >> Well, big Microsoft announcement. If we go back a few years before Nadella took over, Ballmer was a great believer in one Microsoft. They bought Nokia, they were looking at putting Windows into everything, it was a Windows led, one Microsoft organization. And a lot of ambitious ideas were cut off because they didn't get the sign off by, for example, the Windows group. Nadella's first action, and I actually was there, was to announce Office on the iPhone. A major, major thing that had been proposed for a long time was being held up internally. And now he's gone even further. The focus, clear focus of Microsoft is on the cloud, you know 50% plus CAGR on the cloud, Office 365 CAGR 41% and AI, focusing on AI and obviously the intelligent age as well. So Windows 10, Myerson, the leader there, is out, 2% CAGR, he missed his one billion Windows target, by a long way, something like 50%. Windows functionality is being distributed, essentially, across the whole of Microsoft. So hardware is taking the Xbox and the Surface. Windows server itself is going to the cloud. So, big change from the historical look of Microsoft, but, a trimming down of the organization and a much clearer focus on the key things driving Microsoft's fantastic increase in net worth. >> So Microsoft retooling to take advantage and be more relevant, sustain it's relevance in the new era of computing. Once again, this has been a Wikibon Action Item Quick Take. (soft electronic music)

>> Hi, I'm Peter Burris, and welcome to a Wikibon Action Item Quick Take. Jim Kobielus, lots going on in the world of AI and storage. If we think about what happened in storage over the years, it used to be for disc space, get data into a persistent state, and for some of the flash base, it's get data out faster. What happened this week between Pure and NVIDIA to make it easier to get data out faster, especially for AI applications? >> Yeah Peter, this week at NVIDIA's annual conference, GPU technology conference, they announced a partnership with Pure Storage. In fact they released a jointly developed product called AIRI...A-I-R-I standing for AI Ready Infrastructure. What's significant about AIRI is that it is a... Well, I'll tell you years ago, I'm showing my age there was this constant well of data warehousing appliance, a pre-bundled, pre-integrated assembly of storage and compute and software for specific workloads. Though, I wouldn't use the term appliance here, it's a similar concept. In the AI space, there's a need for pre-integrated storage and compute devices...racks...for training workloads and other core, very compute and very data-intensive workloads for AI And that's what the Pure Storage NVIDIA AIRI is all about. It includes Pure Storage's Flashblade storage technology, plus four NVIDIA DCX supercomputers that are running the latest GPUs, the Tesla V100. As well as providing a fast interconnect of NVIDIA's. Plus, also bundling software, NVIDIA's AI frame was from modeling, there's a management tool from Pure Storage. What this is, this is a harbinger of what we expect, and Wikibon will be a broader range from these vendors and others of pre-built optimized AI storage products for premises based deployment, for hyperquads, really for complex AI pipelines involving data... Scientist data, engineers and others. We're very excited about this particular product, we think it has great potential and we believe there's a lot of pent-up demand for these kinds of pre-built hardware products. And that, in many ways, was by far the most significant story in the AI space this week. >> All right, so this has been...thanks very much for that Jim. So, more to come, moving more compute closer to the data. Part of a bigger trend. This has been a Wikibon Action Item Quick Take. >> (smooth techno music)

>> Hi, I'm Peter Burris, and welcome to another Wikibon Action Item. (funky electronic music) This was a very interesting week in the tech industry, specifically because IBM's Think Conference aggregated in a large number of people. Now, The CUBE was there. Dave Vellante, John Furrier, and myself all participated in somewhere in the vicinity of 60 or 70 interviews with thought leaders in the industry, including a number of very senior IBM executives. The reason why this becomes so important is because IBM made a proposal to the industry about how some of the digital disruption that the market faces is likely to unfold. The normal approach or the normal mindset that people have used is that startups, digital native companies were going to change the way that everything was going to operate, and the dinosaurs were going to go by the wayside. IBM's interesting proposal is that the dinosaurs actually are going to learn to dance, utilizing or playing on a book title from a number of years ago. And the specific argument was laid out by Ginni Rometty in her keynote, when she said that there are number of factors that are especially important here. Factor number one is that increasingly, businesses are going to recognize that the role that their data plays in competition is on the ascending. It's getting more important. Now, this is something that Wikibon's been arguing for quite some time. In fact, we have said that the whole key to digital disruption and digital business is to acknowledge the difference between business and digital business, is the role that data and data assets play in your business. So we have strong agreement there. But on top of that, Ginni Rometty made the observation that 80% of the data that could be accessed and put the work in business has not yet been made available to the new activities, the new processes that are essential to changing the way customers are engaged, businesses operate, and overall change and disruption occurs. So her suggestion is that that 80%, that vast amount of data that could be applied that's not being tapped, is embedded deep within the incumbents. And so the core argument from IBM is that the incumbent companies, not the digital natives, not the startups, but the incumbent companies are poised to make a significant, to have a significant role in disrupting how markets operate, because of the value of their data that hasn't currently been put to work and made available to new types of work. That was the thesis that we heard this week, and that's what we're going to talk about today. Are the incumbent really going to strike back? So Dave Vellante, let me start with you. You were at Think, you heard the same type of argument. What did you walk away with? >> So when I first heard the term incumbent disruptors, I was very skeptical, and I still am. But I like the concept and I like it a lot. So let me explain why I like it and why I think there's some real challenges. If I'm a large incumbent global 2,000, I'm not going to just roll over because the world is changing and software is eating my world. Rather what I'm going to do is I'm going to use my considerable assets to compete, and so that includes my customers, my employees, my ecosystem, the partnerships that I have there, et cetera. The reason why I'm skeptical is because incumbents aren't organized around their data assets. Their data assets are stovepipe, they're all over the place. And the skills to leverage that data value, monetize that data, understand the contribution that data makes toward monetization, those skills are limited. They're bespoke and they're very narrow. They're within lines of business or divisions. So there's a huge AI gap between the true digital business and an incumbent business. Now, I don't think all is lost. I think a lot of strategies can work, from M&A to transformation projects, joint ventures, spin-offs. Yeah, IBM gave some examples. They put up Verizon and American Airlines. I don't see them yet as the incumbent disruptors. But then there was another example of IBM Maersk doing some very interesting and disrupting things, Royal Bank of Canada doing some pretty interesting things. >> But in a joint venture forum, Dave, to your point, they specifically set up a joint venture that would be organized around this data, didn't they? >> Yes, and that's really the point I'm trying to make. All is not lost. There are certain things that you can do, many things that you can do as an incumbent. And it's really game on for the next wave of innovation. >> So we agree as a general principle that data is really important, David Floyer. And that's been our thesis for quite some time. But Ginni put something out there, Ginni Rometty put something out there. My good friend, Ginni Rometty, put something out there that 80% of the data that could be applied to disruption, better customer engagement, better operations, new markets, is not being utilized. What do we think about that? Is that number real? >> If you look at the data inside any organization, there's a lot of structured data. And that has better ability to move through an organization. Equally, there's a huge amount of unstructured data that goes in emails. It goes in voicemails, it goes in shared documents. It goes in diagrams, PowerPoints, et cetera, that also is data which is very much locked up in the way that Dave Vellante was talking about, locked up in a particular process or in a particular area. So is there a large amount of data that could be used inside an organization? Is it private, is it theirs? Yes, there is. The question is, how do you tap that data? How do you organize around that data to release it? >> So this is kind of a chicken and egg kind of a problem. Neil Raden, I'm going to turn to you. When we think about this chicken and egg problem, the question is do we organize in anticipation of creating these assets? Do we establish new processes in anticipation of creating these data assets? Or do we create the data assets first and then re-institutionalize the work? And the reason why it's a chicken and egg kind of problem is because it takes an enormous amount of leadership will to affect the way a business works before the asset's in place. But it's unclear that we're going to get the asset that we want unless we affect the reorganization, institutionalization. Neil, is it going to be a chicken? Is it going to be the egg? Or is this one of the biggest problems that these guys are going to have? >> Well, I'm a little skeptical about this 80% number. I need some convincing before I comment on that. But I would rather see, when David mentioned the PowerPoint slides or email or that sort of thing, I would rather see that information curated by the application itself, rather than dragged out in broad data and reinterpreted in something. I think that's very dangerous. I think we saw that in data warehousing. (mumbling) But when you look at building data lakes, you throw all this stuff into a data lake. And then after the fact, somebody has to say, "Well, what does this data mean?" So I find it kind of a problem. >> So Jim Kobielus, a couple weeks ago Microsoft actually introduced a technology or a toolkit that could in fact be applied to move this kind of advance processing for dragging value out of a PowerPoint or a Word document or something else, close and proximate to the application. Is that, I mean, what Neil just suggested I think is a very, very good point. Are we going to see these kinds of new technologies directly embedded within applications to help users narrowly, but businesses more broadly, lift that information out of these applications so it can be freed up for other uses? >> I think yeah, on some level, Peter, this is a topic called dark data. It's been discussed in data management circles for a long time. The vast majority, I think 75 to 80% is the number that I see in the research, is locked up in terms of it's not searchable, it's not easily discoverable. It's not mashupable, I'm making up a word. But the term mashup hasn't been used in years, but I think it's a good one. What it's all about is if we want to make the most out of our incumbent's data, then we need to give the business, the business people, the tools to find the data where it is, to mash it up into new forms and analytics and so forth, in order to monetize it and sell it, make money off of it. So there are a wide range of data discovery and other tools that support a fairly self-service combination and composition of composite data object. I don't know that, however, that the culture of monetizing existing dataset and pulling dark data into productized forms, I don't think that's taken root in any organization anywhere. I think that's just something that consultants talk about as something that gee, should be done, but I don't think it's happening in the real world. >> And I think you're probably correct about that, but I still think Neil raised a great point. And I would expect, and I think we all believe, that increasingly this is not going to come as a result of massive changes in adoption of new data science, like practices everywhere, but an embedding of these technologies. Machine learning algorithms, approaches to finding patterns within application data, in the applications themselves, which is exactly what Neil was saying. So I think that what we're going to see, and I wanted some validation from you guys about this, is increasingly tools being used by application providers to reveal data that's in applications, and not open source, independent tool chains that then ex-post-facto get applied to all kinds of different data sources in an attempt for the organization to pull the stuff out. David Floyer, what do you think? >> I agree with you. I think there's a great opportunity for the IT industry in this area to put together solutions which can go and fit in. On the basis of existing applications, there's a huge amount of potential, for example, of ERP systems to link in with IOT systems, for example, and provide a data across an organization. Rather than designing your own IOT system, I think people are going to buy-in pre-made ones. They're going to put the devices in, the data's going to come in, and the AI work will be done as part of that, as part of implementing that. And right across the board, there is tremendous opportunity to improve the applications that currently exist, or put in new versions of applications to address this question of data sharing across an organization. >> Yeah, I think that's going to be a big piece of what happens. And it also says, Neil Raden, something about whether or not enormous machine learning deities in the sky, some of which might start with the letter W, are going to be the best and only way to unlock this data. Is this going to be something that, we're suggesting now that it's something that's going to be increasingly-distributed closer to applications, less invasive and disruptive to people, more invasive and disruptive to the applications and the systems that are in place. And what do you think, Neil? Is that a better way of thinking about this? >> Yeah, let me give you an example. Data science the way it's been practiced is a mess. You have one person who's trying to find the data, trying to understand the data, complete your selection, designing experiments, doing runs, and so forth, coming up with formulas and then putting them in the cluster with funny names so they can try to remember which one was which. And now what you have are a number of software companies who've come up with brilliant ways of managing that process, of really helping the data science to create a work process in curating the data and so forth. So if you want to know something about this particular model, you don't have to go to the person and say, "Why did you do that model? "What exactly were you thinking?" That information would be available right there in the workbench. And I think that's a good model for, frankly, everything. >> So let's-- >> Development pipeline toolkits. That's a hot theme. >> Yeah, it's a very hot theme. But Jim, I don't think you think but I'm going to test it. I don't think we're going to see AI pipeline toolkits be immediately or be accessed by your average end user who's putting together a contract, so that that toolkit or so that data is automatically munched and ingested or ingested and munched by some AI pipeline. This is going to happen in an application. So the person's going to continue to do their work, and then the tooling will or will not grab that information and then combine it with other things through the application itself into the pipeline. We got that right? >> Yeah, but I think this is all being, everything you described is being embedded in applications that are making calls to backend cloud services that have themselves been built by data scientists and exposed through rest APIs. Steve, Peter, everything you're describing is coming to applications fairly rapidly. >> I think that's a good point, but I want to test it. I want to test that. So Ralph Finos, you've been paying a lot of attention during reporting season to what some of the big guys are saying on some of their calls and in some of their public statements. One company in particular, Oracle, has been finessing a transformation, shall we say? What are they saying about how this is going as we think about their customer base, the transformation of their customer base, and the degree to which applications are or are not playing a role in those transformations? >> Yeah, I think in their last earnings call a couple days ago that the point that they were making around the decline and the-- >> Again, this is Oracle. So in Oracle's last earnings call, yeah. >> Yeah, I'm sorry, yeah. And the decline and the revenue growth rate in the public cloud, the SAS end of their business, was a function really of a slowdown of the original acquisitions they made to kind of show up as being a transformative cloud vendor, and that are basically beginning to run out of gas. And I think if you're looking at marketing applications and sales-related applications and content-type of applications, those are kind of hitting a natural high of growth. And I think what they were saying is that from a migration perspective on ERP, that that's going to take a while to get done. They were saying something like 10 or 15% of their customer base had just begun doing some sort of migration. And that's a data around ERP and those kinds of applications. So it's a long slog ahead of them, but I'd rather be in their shoes, I think, for the long run than trying to kind of jazz up in the near-term some kind of pseudo-SAS cloud growth based on acquisition and low-lying fruit. >> Yeah, because they have a public cloud, right? I mean, at least they're in the game. >> Yeah, and they have to show they're in the game. >> Yeah, and specifically they're talking about their applications as clouds themselves. So they're not just saying here's a set of resources that you can build, too. They're saying here's a set of SAS-based applications that you can build around. >> Dave: Right. Go ahead, Ralph, sorry. >> Yeah, yeah. And I think the notion there is the migration to their ERP and their systems of record applications that they're saying, this is going to take a long time for people to do that migration because of complexity in process. >> So the last point, or Dave Vellante, did you have a point you want to make before I jump into a new thought here? >> I just compare and contrast IBM and Oracle. They have public clouds, they have SAS. Many others don't. I think this is a major different point of differentiation. >> Alright, so we've talked about whether or not this notion of data as a source of value's important, and we agree it is. We still don't know whether or not 80% is the right number, but it is some large number that's currently not being utilized and applied to work differently than the data currently is. And that likely creates some significant opportunities for transformation. Do we ultimately think that the incumbents, again, I mention the chicken and the egg problem. Do we ultimately think that the incumbents are... Is this going to be a test of whether or not the incumbents are going to be around in 10 years? The degree to which they enact the types of transformation we thought about. Dave Vellante, you said you were skeptical. You heard the story. We've had the conversation. Will incumbents who do this in fact be in a better position? >> Well, incumbents that do take action absolutely will be in a better position. But I think that's the real question. I personally believe that every industry is going to get disrupted by digital, and I think a lot of companies are not prepared for this and are going to be in deep trouble. >> Alright, so one more thought, because we're talking about industries overall. There's so many elements we haven't gotten to, but there's one absolute thing I want to talk about. Specifically the difference between B2C and B2B companies. Clearly the B2C industries have been disrupted, many of them pretty significantly, over the last few years. Not too long ago, I have multiple not-necessarily-good memories of running the aisles of Toys R Us sometime after 10 o'clock at night, right around December 24th. I can't do that anymore, and it's not because my kids are grown. Or I won't be able to do that soon anymore. So B2C industries seem to have been moved faster, because the digital natives are able to take advantage of the fact that a lot of these B2C industries did not have direct and strong relationships with those customers. I would posit that a lot of the B2B industries are really where the action's going to take. And the kind of way I would think about it, and David Floyer, I'll turn to you first. The way I would think about it is that in the B2C world, it's new markets and new ways of doing things, which is where the disruption's going to take place. So more of a substitution as opposed to a churn. But in the B2B markets, it's disrupting greater efficiencies, greater automation, greater engagement with existing customers, as well as finding new businesses and opportunities. What do you think about that? >> I think the B2B market is much more stable. Relationships, business relationships, very, very important. They take a long time to change. >> Peter: But much of that isn't digital. >> A lot of that is not digital. I agree with that. However, I think that the underlying change that's happening is one of automation. B2B are struggling to put into place automation with robots, automation everywhere. What you see, for example, in Amazon is a dedication to automation, to making things more efficient. And I think that's, to me, the biggest challenges, owning up to the fact that they have to change their automation, get themselves far more efficient. And if they don't succeed in doing that, then their ability to survive or their likelihood of being taken over with a reverse takeover becomes higher and higher and higher. So how do you go about that level, huge increase in automation that is needed to survive, I think is the biggest question for B2B players. >> And when we think about automation, David Floyer, we're not talking about the manufacturing arms or only talking about the manufacturing arms. We're talking about a lot of new software automation. Dave Vellante, Jim Kobielus, RPA is kind of a new thing. Dave, we saw some interesting things at Think. Bring us up to speed quickly on what the community at Think was talking about with RPA. >> Well, I tell you. There were a lot of people in financial services, which is IBM's stronghold. And they're using software robots to automate a lot of the backend stuff that humans were doing. That's a major, major use case. I would say 25 to 30% of the financial services organizations that I talked to had active RPA projects ongoing at the moment. I don't know. Jim, what are your thoughts? >> Yeah, I think backend automation is where B2B disruption is happening. As the organizations are able to automate more of their backend, digitize more of their backend functions and accelerate them and improve the throughput of transactions, are those that will clean up. I think for the B2C space, it's the frontend automation of the digitalization of the engagement channels. But RPA is essentially a key that's unlocking backend automation for everybody, because it allows more of the frontend business analysts and those who are not traditionally BPM, or business process re-engineering professionals, to begin to take standard administrative processes and begin to automate them from, as it were, the outside-in in a greater way. So I think RPA is a secret key for that. I think we'll see some of the more disruptive organizations, businesses, take RPA and use it to essentially just reverse-engineer, as it were, existing processes, but in an automated fashion, and drive that improvement but in the backend by AI. >> I just love the term software robots. I just think that that's, I think that so strongly evokes what's going to happen here. >> If I could add, I think there's a huge need to simplify that space. The other thing I witnessed at IBM Think is it's still pretty complicated. It's still a heavy lift. There's a lot of big services component to this, which is probably why IBM loves it. But there's a massive market, I think, to simplify the adoption or RPA. >> I completely agree. We have to open the aperture as well. Again, the goal is not to train people new things, new data science, new automation stuff, but to provide tools and increasingly embed those tools into stuff that people are already using, so that the disruption and the changes happen more as a consequence of continuing to do what the people do. Alright, so let's hit the action item we're on, guys. It's been a great conversation. Again, we haven't talked about GDPR. We haven't talked about a wide array of different factors that are going to be an issue. I think this is something we're going to talk about. But on the narrow issue of can the disruptors strike back? Neil Raden, let's start with you. Neil Raden, action item. >> I've been saying since 1975 that I should be hanging around with a better class of people, but I do spend a lot of time in the insurance industry. And I have been getting a consensus that in the next five to 10 years, there will no longer be underwriters for claims adjustments. That business is ready for massive, massive change. >> And those are disruptors, largely. Jim Kobielus, action item. >> Action item. In terms of business disruption, is just not to imagine that because you were the incumbent in the past era in some solution category that's declining, that that automatically guarantees you, that makes your data fit for seizing opportunities in the future. As we've learned from Blockbuster Video, the fact that they had all this customer data didn't give them any defenses against Netflix coming along and cleaning their coffin, putting them out of business. So the next generation of disruptor will not have any legacy data to work from, and they'll be able to work miracles because they made a strategic bet on some frontend digital channel that made all the difference. >> Ralph Finos, action item. >> Yeah, I think there's a notion here of siege mentality. And I think the incumbents are in the castle walls, and the disruptors are outside the castle walls. And sometimes the disruptors, you know, scale the walls. Sometimes they don't. But I think being inside the walls is a long-run tougher thing to be at. >> Dave Vellante, action item. >> I want to pick up on something Neil said. I think it's alluring for some of these industries, like insurance and financial services and healthcare, even parts of government, that really haven't been disrupted in a huge way yet to say, "Well, I'll wait and I'll see what happens." I think that's a huge mistake. I think you have to start immediately thinking about strategies, particularly around your data, as we talked about earlier. Maybe it's M&A, maybe it's joint ventures, maybe it's spinning out new companies. But the time is past where you should be acting. >> David Floyer, action item. >> I think that it's easier to focus on something that you can actually do. So my action item is that the focus of most B2B companies should be looking at all of their processes and incrementally automating them, taking out the people cost, taking out the cost, other costs, automating those processes as much as possible. That, in my opinion, is the most likely path to being in the position that you can continue to be competitive. Without that focus, it's likely that you're going to be disrupted. >> Alright. So the one thing I'll say about that, David, is when I think you say people cost I think you mean the administrative cost associated with people. >> And people doing things, automating jobs. >> Alright, so we have been talking here in today's Wikibon Action Item about the question, will the incumbents be able to strike back? The argument we heard at IBM Think this past week, and this is the third week of March, was that data is an asset that can be applied to significantly disrupt industries, and that incumbents have a lot of data that hasn't been bought into play in the disruptive flow. And IBM's argument is that we're going to see a lot of incumbents start putting their data into play, more of their data assets into play. And that's going to have a significant impact ultimately on industry structure, customer engagement, the nature of the products and services that are available over the course of the next decade. We agree. We generally agree. We might nitpick about whether it's 80%, whether it's 60%. But in general, the observation is an enormous amount of data that exists within a large company, that's related to how they conduct business, is siloed and locked away and is used once and is not made available, is dark and is not made available for derivative uses. That could, in fact, lead to significant consequential improvements in how a business's transaction costs are ultimately distributed. Automation's going to be a big deal. David Floyer's mentioned this in the past. I'm also of the opinion that there's going to be a lot of new opportunities for revenue enhancement and products. I think that's going to be as big, but it's very clear that to start it makes an enormous amount of sense to take a look at where your existing transaction costs are, where existing information asymmetries exist, and see what you can do to unlock that data, make it available to other processes, and start to do a better job of automating local and specific to those activities. And we generally ask our clients to take a look at what is your value proposition? What are the outcomes that are necessary for that value proposition? What activities are most important to creating those outcomes? And then find those that, by doing a better job of unlocking new data, you can better automate those activities. In general, our belief is that there's a significant difference between B2C and B2B businesses. Why? Because a lot of B2C businesses never really had that direct connection, therefore never really had as much of the market and customer data about what was going on. A lot of point-of-sale perhaps, but not a lot of other types of data. And then the disruptors stepped in and created direct relationships, gathered that data and were able to rapidly innovate products and services that served consumers differently. Where a lot of that new opportunity exists is in the B2B world. And here's where the real incumbents are going to start flexing their muscles over the course of the next decade, as they find those opportunities to engage differently, to automate existing practices and activities, change their cost model, and introduce new approaches to operating that are cloud-based, blockchain-based, data-based, based on data, and find new ways to utilize their people. If there's one big caution we have about this, it's this. Ultimately, the tooling is not broadly mature. The people necessary to build a lot of these tools are increasingly moving into the traditional disruptors, the legacy disruptors if we will. AWS, Netflix, Microsoft, companies more along those lines. That talent is very dear still in the industry, and it's going to require an enormous effort to bring those new types of technologies that can in fact liberate some of this data. We looked at things like RPA, robot process automation. We look at the big application providers to increasingly imbue their products and services with some of these new technologies. And ultimately, paradoxically perhaps, we look for the incumbent disruptors to find ways to disrupt without disrupting their own employees and customers. So embedding more of these new technologies in an ethical way directly into the systems and applications that serve people, so that the people face minimal changes to learning new tricks, because the systems themselves have gotten much more automated and much more... Are able to learn and evolve and adjust much more rapidly in a way that still corresponds to the way people do work. So our action item. Any company in the B2B space that is waiting for data to emerge as an asset in their business, so that they can then do all the institutional, re-institutionalizing of work and reorganizing of work and new types of investment, is not going to be in business in 10 years. Or it's going to have a very tough time with it. The big challenge for the board and the CIO, and it's not successfully been done in the past, at least not too often, is to start the process today without necessarily having access to the data, of starting to think about how the work's going to change, think about the way their organization's going to have to be set up. This is not business process re-engineering. This is organizing around future value of data, the options that data can create, and employ that approach to start doing local automation, serve customers, and change the way partnerships work, and ultimately plan out for an extended period of time how their digital business is going to evolve. Once again, I want to thank David Floyer here in the studio with me. Neil Raden, Dave Vellante, Ralph Finos, Jim Kobielus remote. Thanks very much guys. For all of our clients, once again this has been a Wikibon Action Item. We'll talk to you again. Thanks for watching. (funky electronic music)

>> Hi I'm Peter Burris, and welcome to another Wikibon Action Item Quick Take. Dave Vellante, interesting community event next week. What's going on? >> So Infinidat is a company that was started by Moshe Yanai. He invented symmetrics... Probably the single most important storage product of all time. At any rate, he started this new company Infinidat. They tend to do things differently. They're a one product company, but Tuesday March 27th, they're extending their portfolio pretty dramatically. We're going to be covering that. We have a crowd chat going on. It's, again Tuesday March 27th, 10:30 eastern time. Crowdchat.net/infinichat. Check it out. >> Great. That's been our Wikibon Action Item Quick Take. Talk to you soon. (upbeat music)