>>from around the globe. It's the >>cube with digital >>coverage of IBM >>Think 2021 brought to you >>by IBM. Welcome back everyone to the cube coverage of IBM Think 2021 virtual um john for your host of the cube. Our next guest taylor Holloway. Chief technology officer at advent one. Tyler welcome to the cube from down under in Australia and we're in Palo alto California. How are you? >>Well thanks john thanks very much. Glad to be glad to be on here. >>Love love the virtual cube of the virtual events. We can get to talk to people really quickly with click um great conversation here around hybrid cloud, multi cloud and all things software enterprise before we get started. I wanna take a minute to explain what you guys do at advent one. What's the main focus? >>Yeah. So look we have a lot of customers in different verticals. Um so you know generally what we provide depends on the particular industry the customers in. But generally speaking we see a lot of demand for operational efficiency, helping our clients tackle cyber security risks, adopt cloud and set them up to modernize the applications. >>And this is this has been a big wave coming in for sure with you know, cloud and scale. So I gotta ask you, what are the main challenges that you guys are solvent for your customers um and how are you helping them overcome come that way and transformative innovative way? >>Yeah, look, I think helping our clients um improve their security posture is a big one. We're finding as well that our customers are gaining a lot of operational efficiency by adopting sort of open source technology red huts an important partner of ours as his IBM um and we're seeing them sort of move away from some more proprietary solutions. Automation is a big focus for us as well. We've had some great outcomes with our clients or helping them automate um and you know, to live up um you know the stand up and data operations of environments a lot quickly a lot more easily and uh and to be able to sort of apply some standards across multiple sort of areas of their I. T. Estate. >>What are some of the solutions that you guys are doing with IBM's portfolio on the infrastructure side, you got red hat, you've got a lot of open source stuff to meet the needs of clients. What do you mean? What's the mean? >>Uh Yeah I think on the storage side will probably help our clients sort of tackle the expanding data in structured and particularly unstructured data they're trying to take control of so you know, looking at spectrum scale and those type of products from an audio perspective for unstructured data is a good example. And so they're flush systems for more block storage and more run of the mill sort of sort of environments. We have helped our clients consolidate and modernize on IBM power systems. Having Red Hat is both a Lynx operating system and having open shift as a container platform. Um really helps there. And Red Hat also provides management overlay, which has been great on what we do with IBM power systems. We've been working on a few different sort of use cases on power in particular. More recently, SAP Hana is a big one where we've had some success with our clients migrating Muhanna on to onto IBM power systems. And we've also helped our customers, you know, improve some um some environments on the other end of the side, such as IBM I, we still have a large number of customers with, with IBM I and and you know how do we help them? You know some of them are moving to cloud in one way or another others are consuming some kind of IRS and we can sort of wrap around a managed service to to help them through. >>So I gotta ask you the question, you know U C T. Oh you played a lot of technologies kubernetes just become this lingua franca for this kind of like I'll call a middleware kind of orchestration layer uh containers. Also you're awesome but I gotta ask you when you walk into a client's environment you have to name names but you know usually you see kind of two pictures man, they need some serious help or they got their act together. So either way they're both opportunities for Hybrid cloud. How do you how do you how do you evaluate the environment when you go in, when you walk into those two scenarios? What goes through your mind? What some of the conversations that you guys have with those clients. Can you take me through a kind of day in the life of both scenarios? The ones that are like I can't get the job done, I'm so close in on the right team and the other ones, like we're grooving, we're kicking butt. >>Yeah. So look, let's start well, I supposed to start off with you try and take somewhat of a technology agnostic view and just sort of sit down and listen to what they're trying to achieve, how they're going for customers who have got it. You know, as you say, all nailed down things are going really well. Um it's just really understanding what what can we do to help. Is there an opportunity for us to help at all like there? Um, you know, generally speaking, there's always going to be something and it may be, you know, we don't try and if someone is going really well, they might just want someone to help with a bespoke use case or something very specific where they need help. On the other end of the scale where a customer is sort of pretty early on and starting to struggle. We generally try and help them not boil the ocean at once. Just try and get some winds, pick some key use cases, you know, deliver some value back and then sort of growing from there rather than trying to go into a customer and trying to do everything at once tends to be a challenge. Just understand what the priorities are and help them get going. >>What's the impact been for red hat? Um, in your customer base, a lot of overlap. Some overlap, no overlap coming together. What's the general trend that you're seeing? What's the reaction been? >>Yeah I think it's been really good. Obviously IBM have a lot of focus on cloud packs where they're bringing their software on red hat open shift that will run on multiple clouds. So I think that's one that we'll see a lot more of overtime. Um Also helping customers automate their I. T. Operations with answerable is one we do quite a lot of um and there's some really bespoke use cases we've done with that as well as some standardized one. So helping with day two operations and all that sort of thing. But there's also some really sort of out there things customers have needed to automate that's been a challenge for them and being able to use open source tools to do it has worked really well. We've had some good wins there, >>you know, I want to ask you about the architecture and I'm just some simplify it real. Just for the sake of devops, um you know, segmentation, you got hybrid clouds, take a programmable infrastructure and then you've got modern applications that need to have a I some have said I've even sit on the cube and other broadcast that if you don't have a I you're gonna be at a handicap some machine learning, some data has to be in there. You can probably see ai and mostly everything as you go in and try to architect that out for customers um and help them get to a hybrid cloud infrastructure with real modern application front end with using data. What's what's the playbook? Do you have any best practices or examples you can share or scenarios or visions that you see uh playing >>out? I think you're the first one is obviously making sure customers data is in the right place. So if they might be wanting to use um some machine learning in one particular cloud provider and they've got a lot of their applications and data in another, you know, how do we help them make it mobile and able to move data from one cloud to another or back into court data center? So there's a lot of that. I think that we spend a lot of time with customers to try and get a right architecture and also how do we make sure it's secure from end to end. So if they're moving things from into multiple one or more public clouds as well as maybe in their own data center, making sure connectivity is all set up properly. All the security requirements are met. So I think we sort of look at it from a from a high level design point of view, we look at obviously what the target state is going to be versus the current state that really take into account security, performance, connectivity or those sort of things to make sure that they're going to have a good result. >>You know, one of the things you mentioned and this comes up a lot of my interviews with partners of IBM is they always comment about their credibility and all the other than the normal stuff. But one of the things that comes out a lot pretty much consistently is their experience in verticals. Uh they have such a track record in verticals and this is where AI and machine learning data has to be very much scoped in on the vertical. You can't generalize and have a general purpose data plane inside of vertically specialized kind of focus. How how do you see that evolving, how does IBM play there with this kind of the horizontally scalable mindset of a hybrid model, both on premise in the cloud, but that's still saying provide that intimacy with the data to fuel the machine learning or NLP or power that ai which seems to be critical. >>Yeah, I think there's a lot of services where you know, public cloud providers are bringing out new services all the time and some of it is pre can and easy to consume. I think what IBM from what I've observed, being really good at is handling some of those really bespoke use cases. So if you have a particular vertical with a challenge, um you know, there's going to be sort of things that are pre can that you can go and consume. But if you need to do something custom that could be quite challenging. How do they sort of build something that could be quite specific for a particular industry and then obviously being able to repeat that afterwards for us, that's obviously something we're very interested in. >>Yeah, tell I love chatting whether you love getting the low down also, people might not know your co author of a book performance guy with IBM Power Systems, So I gotta ask you, since I got you here and I don't mean to put you on the spot, but if you can just share your vision or any kind of anecdotal observation as people start to put together their architecture and again, you know, Beauty's in the eye of the beholder, every environment is different. But still, hybrid, distributed concept is distributed computing. Is there a KPI is there a best practice on as a manager or systems architect to kind of keep an eye on what what good is and how how good becomes better because the day to operations becomes a super important concept. We're seeing some called Ai ops where okay, I'm provisioning stuff out on a hybrid Cloud operational environment. But now day two hits are things happen as more stuff entered into the equation. What's your vision on KPs and management? What to keep tracking? >>Yeah, I think obviously attention to detail is really important to be able to build things properly. A good KPI particularly managed service area that I'm curious that understanding is how often do you actually have to log into the systems that you're managing? So if you're logging in and recitation into servers and all this sort of stuff all the time, all of your automation and configuration management is not set up properly. So, really a good KPI an interesting one is how often do you log into things all the time? If something went wrong, would you sooner go and build another one and shoot the one that failed or go and restore from backup? So thinking about how well things are automated. If things are immutable using infrastructure as code, those are things that I think are really important when you look at, how is something going to be scalable and easy to manage going forward. What I hate to see is where, you know, someone build something and automates it all in the first place and they're too scared to run it again afterwards in case it breaks something. >>It's funny the next generation of leaders probably won't even know like, hey, yeah, taylor and john they had to log into systems back in the day. You know, I mean, I could be like a story they tell their kids. Uh but no, that's a good Metro. This is this automation. So it's on the next level. Let's go the next level automation. Um what's the low hanging fruit for automation? Because you're getting at really the kind of the killer app there, which is, you know, self healing systems, good networks that are programmable but automation will define more value. What's your take? >>I think the main thing is where you start to move from a model of being able to start small and automate individual things which could be patching or system provisioning or anything like that. But what you really want to get to is to be able to drive everything through, get So instead of having a written up paper, change request, I'm going to change your system and all the rest of it. It really should be driven through a pull request and have things through it and and build pipelines to go and go and make a change running in development, make sure it's successful and then it goes and gets pushed into production. That's really where I think you want to get to and you can start to have a lot of people collaborating really well on this particular project or a customer that also have some sort of guard rails around what happens in some level of governance rather than being a free for all. >>Okay, final question. Where do you see event one headed? What's your future plans to continue to be a leader? I. T. Service leader for this guy? BMS Infrastructure portfolio? >>I think it comes down to people in the end, so really making sure that we partner with our clients and to be well positioned to understand what they want to achieve and and have the expertise in our team to bring to the table to help them do it. I think open source is a key enabler to help our clients adopt a hybrid cloud model to sort of touched on earlier uh as well as be able to make use of multiple clouds where it makes sense from a managed service perspective. I think everyone is really considering themselves and next year managed service provider. But what that means for us is to provide a different, differentiated managed service and also have the strong technical expertise to back it up. >>Taylor Holloway, chief technology officer advent one remote videoing in from down under in Australia. I'm john ferrier and Palo alto with cube coverage of IBM thing. Taylor, thanks for joining me today from the cube. >>Thank you very much. >>Okay, cube coverage. Thanks for watching ever. Mhm mm

>>from around the globe. It's the cube with digital >>coverage of IBM >>Think 2021 brought to >>you by IBM. Welcome back everyone to the cube coverage of IBM Think 2021 virtual um john for your host of the cube. Our next guest taylor Holloway. Chief technology officer at advent one. Tyler welcome to the cube from down under in Australia and we're in Palo alto California. How are you? >>Well thanks john thanks very much. Glad to be glad to be on here. >>Love love the virtual cube of the virtual events. We can get to talk to people really quickly with click um great conversation here around hybrid cloud, multi cloud and all things software enterprise before we get started. I wanna take a minute to explain what you guys do at advent one. What's the main focus? >>Yeah. So look we have a lot of customers in different verticals. Um so you know generally what we provide depends on the particular industry the customers in. But generally speaking we see a lot of demand for operational efficiency, helping our clients tackle cyber security risks, adopt cloud and set them up to modernize the applications. >>And this is this has been a big wave coming in for sure with, you know, cloud and scale. So I gotta ask you, what are the main challenges that you guys are solvent for your customers um and how are you helping them overcome come that way and transformative innovative way? >>Yeah, look, I think helping our clients um improve their security posture is a big one. We're finding as well that our customers are gaining a lot of operational efficiency by adopting sort of open source technology. Red Hearts, an important partner of ours is IBM um and we're seeing them sort of move away from some more proprietary solutions. Automation is a big focus for us as well. We've had some great outcomes with our clients or helping them automate um and you know deliver um, you know, the stand up and data operations of environments a lot quickly, a lot more easily. And uh and to be able to sort of apply some standards across multiple sort of areas of their estate. >>What are some of the solutions that you guys are doing with IBM's portfolio in the I. T. Infrastructure side? You got red hat, you got a lot of open source stuff to meet the needs of clients. What do you mean? What's that mean? >>Um Yeah, I think on the storage side will probably help our clients sort of tackle the expanding data in structured and particularly unstructured data they're trying to take control of so, you know, looking at spectrum scale and those type of products from an audio perspective for unstructured data is a good example. And so they're flash systems for more block storage and more run of the mill sort of sort of environments. We have helped our clients consolidate and modernize on IBM Power systems. Having Red Hat is both a UNIX operating system and having I can shift as a container platform really helps there. And Red Hat also provides management overlay, which has been great on what we do with IBM Power systems. We've been working on a few different sort of use cases on power in particular, sort of more recently. Um SAP Hana is a big one where we've had some success with our clients migrating Muhanna on to onto IBM power systems and we've also helped our customers, you know, improve some um some environments on the other end of the side, such as IBM I, we still have a large number of customers with with IBM I and and you know how do we help them? You know some of them are moving to cloud in one way or another others are consuming some kind of IRS and we can sort of wrap around a managed service to to help them through. >>So I gotta ask you the question, you know U. C. T. Oh you played a lot of technology actually kubernetes just become this lingua franca for this kind of like I'll call a middleware kind of orchestration layer uh containers. Obviously you're awesome but I gotta ask you when you walk into a client's environment you have to name names but you know usually you see kind of two pictures man, they need some serious help or they got their act together. So either way they're both opportunities for Hybrid cloud. How do you how do you how do you evaluate the environment when you go in, when you walk into those two scenarios? What goes through your mind? What some of the conversations that you guys have with those clients? Can you take me through a kind of day in the life of both scenarios? The ones that are like I can't get the job done, I'm so close in on the right team and the other ones, like we're grooving, we're kicking butt. >>Yeah. So look, let's start, well, I supposed to start off with you try and take somewhat of a technology agnostic view and just sort of sit down and listen to what they're trying to achieve, how they're going for customers who have got it. You know, as you say, all nailed down things are going really well. Um it's just really understanding what what can we do to help. Is there an opportunity for us to help at all like there? Um, you know, generally speaking, there's always going to be something and it may be, you know, we don't try and if someone is going really well, they might just want someone to help with a bespoke use case or something very specific where they need help. On the other end of the scale where a customer is sort of pretty early on and starting to struggle. We generally try and help them not boil the ocean at once. Just try and get some winds, pick some key use cases, you know, deliver some value back and then sort of growing from there rather than trying to go into a customer and trying to do everything at once tends to be a challenge. Just understand what the priorities are and help them get going. >>What's the impact been for red hat? Um, in your customer base, a lot of overlap. Some overlap, no overlap coming together. What's the general trend that you're seeing? What's the reaction been? >>Yeah I think it's been really good. Obviously IBM have a lot of focus on cloud packs where they're bringing their software on red hat open shift that will run on multiple clouds. So I think that's one that we'll see a lot more of overtime. Um Also helping customers automate their I. T. Operations with answerable is one we do quite a lot of um and there's some really bespoke use cases we've done with that as well as some standardized one. So helping with day two operations and all that sort of thing. But there's also some really sort of out there things customers have needed to automate. That's been a challenge for them and being able to use open source tools to do it has worked really well. We've had some good wins there, >>you know, I want to ask you about the architecture and I'm just some simplify it real just for the sake of devops, um you know, segmentation, you got hybrid clouds, take a programmable infrastructure and then you've got modern applications that need to have a I some have said, I've even said on the cube and other broadcasts that if you don't have a I you're gonna be at a handicap some machine learning, some data has to be in there. You can probably see aI and mostly everything as you go in and try to architect that out for customers um and help them get to a hybrid cloud infrastructure with real modern application front end with using data. What's what's the playbook, do you have any best practices or examples you can share or scenarios or visions that you see uh playing >>out? I think the yeah, the first one is obviously making sure customers data is in the right place. So if they might be wanting to use um some machine learning in one particular cloud provider and they've got a lot of their applications and data in another, you know, how do we help them make it mobile and able to move data from one cloud to another or back into court data center? So there's a lot of that. I think that we spend a lot of time with customers to try and get a right architecture and also how do we make sure it's secure from end to end. So if they're moving things from into multiple one or more public clouds as well as maybe in their own data center, making sure connectivity is all set up properly. All the security requirements are met. So I think we sort of look at it from a from a high level design point of view, we look at obviously what the target state is going to be versus the current state that really take into account security, performance, connectivity or those sort of things to make sure that they're going to have a good result. >>You know, one of the things you mentioned and this comes up a lot of my interviews with partners of IBM is they always comment about their credibility and all the other than the normal stuff. But one of the things that comes out a lot pretty much consistently is their experience in verticals. Uh just have such a track record in verticals and this is where AI and machine learning data has to be very much scoped in on the vertical. You can't generalize and have a general purpose data plane inside of vertically specialized kind of focus. How how do you see that evolving, how does IBM play there with this kind of the horizontally scalable mindset of a hybrid model, both on premise in the cloud, but that's still saying provide that that intimacy with the data to fuel the machine learning or NLP or power that AI, which seems to be critical. >>Yeah, I think there's a lot of services where, you know, public cloud providers are bringing out new services all the time and some of it is pre can and easy to consume. I think what IBM from what I've observed being really good at is handling some of those really bespoke use cases. So if you have a particular vertical with a challenge, um you know, there's going to be sort of things that are pre can that you can go and consume. But if you need to do something custom that could be quite challenging. How do they sort of build something that could be quite specific for a particular industry and then obviously being able to repeat that afterwards for us, that's obviously something we're very interested in. >>Yeah, taylor love chatting, whether you love getting the low down, also, people might not know your co author of a book performance guy with IBM Power Systems, so I gotta ask you, since I got you here and I don't mean to put you on the spot, but if you can just share your vision or any kind of anecdotal observation as people start to put together their architecture and again, you know, Beauty's in the eye of the beholder, every environment is different. But still, hybrid, distributed concept is distributed computing, Is there a KPI is there a best practice on as a manager or systems architect to kind of keep an eye on what what good is and how how good becomes better because the day to operations becomes a super important concept. We're seeing some called Ai ops where Okay, I'm provisioning stuff out on a hybrid Cloud operational environment. But now day two hits are things happen as more stuff entered into the equation. What's your vision on KPs and management? What to keep >>tracking? Yeah, I think obviously attention to detail is really important to be able to build things properly. A good KPI particularly managed service area that I'm curious that understanding is how often do you actually have to log into the systems that you're managing? So if you're logging in and recitation into servers and all this sort of stuff all the time, all of your automation and configuration management is not set up properly. So, really a good KPI an interesting one is how often do you log into things all the time if something went wrong, would you sooner go and build another one and shoot the one that failed or go and restore from backup? So thinking about how well things are automated. If things are immutable using infrastructure as code, those are things that I think are really important when you look at, how is something going to be scalable and easy to manage going forward. What I hate to see is where, you know, someone build something and automated all in the first place and they're too scared to run it again afterwards in case it breaks something. >>It's funny the next generation of leaders probably won't even know like, hey, yeah, taylor and john they had to log into systems back in the day. You know, I mean, I could be like a story they tell their kids. Uh but no, that's a good metric. This is this automation. So it's on the next level. Let's go the next level automation. Um what's the low hanging fruit for automation? Because you're getting at really the kind of the killer app there which is, you know, self healing systems, good networks that are programmable but automation will define more value. >>What's your take? I think the main thing is where you start to move from a model of being able to start small and automate individual things which could be patching or system provisioning or anything like that. But what you really want to get to is to be able to drive everything through. Get So instead of having a written up paper, change request, I'm going to change your system and all the rest of it. It really should be driven through a pull request and have things through it and and build pipelines to go and go and make a change running in development, make sure it's successful and then it goes and gets pushed into production. That's really where I think you want to get to and you can start to have a lot of people collaborating really well on this particular project or a customer that also have some sort of guard rails around what happens in some level of governance rather than being a free for >>all. Okay, final question. Where do you see event one headed? What's your future plans to continue to be a leader? I. T. Service by leader for this guy? BMS infrastructure portfolio? >>I think it comes down to people in the end, so really making sure that we partner with our clients and to be well positioned to understand what they want to achieve and and have the expertise in our team to bring to the table to help them do it. I think open source is a key enabler to help our clients adopt a hybrid cloud model to sort of touched on earlier as well as be able to make use of multiple clouds where it makes sense From a managed service perspective. I think everyone is really considering themselves next year managed service provider, but what that means for us is to provide a different, differentiated managed service and also have the strong technical expertise to back it up. >>Taylor Holloway, chief technology officer advent one remote videoing in from down under in Australia. I'm john ferrier and Palo alto with cube coverage of IBM thing. Taylor, thanks for joining me today from the cube. >>Thank you very much. >>Okay, cube coverage. Thanks for watching ever. Mhm

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

[Voiceover] - Live from Washington D.C. It's theCUBE covering dot next conference. Brought to you by Nutanix. >> Welcome back to the district everybody. This is Nutanix NEXTconf, hashtag NEXTconf. And this is theCUBE, the leader in live tech coverage. Stephanie Chiras is here. She's the Vice President of IBM Power Systems Offering Management, and she's joined by Raja Mukhopadhyay who is the VP of Product Management at Nutanix. Great to see you guys again. Thanks for coming on. >> Yeah thank you. Thanks for having us. >> So Stephanie, you're welcome, so Stephanie I'm excited about you guys getting into this whole hyper converged space. But I'm also excited about the cognitive systems group. It's kind of a new play on power. Give us the update on what's going on with you guys. >> Yeah so we've been through some interesting changes here. IBM Power Systems, while we still maintain that branding around our architecture, from a division standpoint we're now IBM Cognitive Systems. We've been through a change in leadership. We have now Senior Vice President Bob Picciano leading IBM Cognitive Systems, which is foundationally built upon the technology that's comes from Power Systems. So our portfolio remains IBM Power Systems, but really what it means is we've set our sights on how to take our technology into really those cognitive workloads. It's a focus on clients going to the cognitive era and driving their business into the cognitive era. It's changed everything we do from how we deliver and pull together our offerings. We have offerings like Power AI, which is an offering built upon a differentiated accelerated product with Power technology inside. It has NVIDIA GPU's, it has NVLink capability, and we have all the optimized frameworks. So you have Caffe, Torch, TensorFlow, Chainer, Theano. All of those are optimized for the server, downloadable right in a binary. So it's really about how do we bring ease of use for cognitive workloads and allow clients to work in machine learning and deep learning. >> So Raja, again, part of the reason I'm so excited is IBM has a $15 billion analytics business. You guys talk, you guys talked to the analysts this morning about one of the next waves of workloads is this sort of data oriented, AI, machine learning workloads. IBM obviously has a lot of experience in that space. How did this relationship come together, and let's talk about what it brings to customers. >> It was all like customer driven, right? So all our customers they told us that, look Nutanix we have used your software to bring really unprecedented levels of like agility and simplicity to our data center infrastructure. But, you know, they run at certain sets of workloads on, sort of, non IBM platforms. But a lot of mission critical applications, a lot of the, you know, the cognitive applications. They want to leverage IBM for that, and they said, look can we get the same Nutanix one click simplicity all across my data center. And that is a promise that we see, can we bring all of the AHV goodness that abstracts the underlying platform no matter whether you're running on x86, or your cognitive applications, or your mission critical applications on IBM power. You know, it's a fantastic thing for a joint customer. >> So Stephanie come on, couldn't you reach somewhere into the IBM portfolio and pull out a hyper converged, you know, solution? Why Nutanix? >> Clients love it. Look what the hyper converged market is doing. It's growing at incredible rates, and clients love Nutanix, right? We see incredible repurchases around Nutanix. Clients buy three, next they buy 10. Those repurchase is a real sign that clients like the experience. Now you can take that experience, and under the same simplicity and elegance right of the Prism platform for clients. You can pull in and choose the infrastructure that's best for your workload. So I look at a single Prism experience, if I'm running a database, I can pull that onto a Power based offering. If I'm running a BDI I can pull that onto an alternative. But I can now with the simplicity of action under Prism, right for clients who love that look and feel, pick the best infrastructure for the workloads you're running, simply. That's the beauty of it. >> Raja, you know, Nutanix is spread beyond the initial platform that you had. You have Supermicro inside, you've got a few OEMs. This one was a little different. Can you bring us inside a little bit? You know, what kind of engineering work had to happen here? And then I want to understand from a workload perspective, it used to be, okay what kind of general purpose? What do you want on Power, and what should you say isn't for power? >> Yeah, yeah, it's actually I think a power to, you know it speaks to the, you know, the power of our engineering teams that the level of abstraction that they were able to sort of imbue into our software. The transition from supporting x86 platforms to making the leap onto Power, it has not been a significant lift from an engineering standpoint. So because the right abstractions were put in from the get go. You know, literally within a matter of mere months, something like six to eight months, we were able to have our software put it onto the IBM power platform. And that is kind of the promise that our customers saw that look, for the first time as they are going through a re-platforming of their data center. They see the power in Nutanix as software to abstract all these different platforms. Now in terms of the applications that, you know, they are hoping to run. I think, you know, we're at the cusp of a big transition. If you look at enterprise applications, you could have framed them as systems of record, and systems of engagement. If you look forward the next 10 years, we'll see this big shift, and this new class of applications around systems of intelligence. And that is what a lot-- >> David: Say that again, systems of-- >> Systems of intelligence, right? And that is where a lot of like IBM Power platform, and the things that the Power architecture provides. You know, things around better GPU capabilities. It's going to drive those applications. So our customers are thinking of running both the classical mission critical applications that IBM is known for, but as well as the more sort of forward leaning cognitive and data analytics driven applications. >> So Stephanie, on one hand I look at this just as an extension of what IBM's done for years with Linux. But why is it more, what's it going to accelerate from your customers and what applications that they want to deploy? >> So first, one of the additional reasons Nutanix was key to us is they support the Acropolis platform, which is KVM based. Very much supports our focus on being open around our playing in the Linux space, playing in the KVM space, supporting open. So now as you've seen, throughout since we launched POWER8 back in early 2014 we went Little Endian. We've been very focused on getting a strategic set of ISV's ported to the platform. Right, Hortonworks, MongoDB, EnterpriseDB. Now it's about being able to take the value propositions that we have and, you know, we're pretty bullish on our value propositions. We have a two x price performance guarantee on MongoDB that runs better on Power than it runs on the alternative competition. So we're pretty bullish. Now for clients who have taken a stance that their data center will be a hyper converged data center because they like the simplicity of it. Now they can pull in that value in a seamless way. To me it's really all about compatibility. Pick the best architecture, and all compatible within your data center. >> So you talked about, six to eight months you were able to do the integration. Was that Open Power that allowed you to do that, was it Little Endian, you know, advancements? >> I think it was a combination of both, right? We have done a lot from our Linux side to be compatible within the broad Linux ecosystem particularly around KVM. That was critical for this integration into Acropolis. So we've done a lot from the bottoms up to be, you know, Linux is Linux is Linux. And just as Raja said, right, they've done a lot in their platform to be able to abstract from the underlying and provide a seamless experience that, you know, I think you guys used the term invisible infrastructure, right? The experience to the client is simple, right? And in a simple way, pick the best, right for the workload I run. >> You talked about systems of intelligence. Bob Picciano a lot of times would talk about the insight economy. And so we're, you're right we have the systems of records, systems of engagement. Systems of intelligence, let's talk about those workloads a little bit. I infer from that, that you're essentially basically affecting outcomes, while the transaction is occurring. Maybe it's bringing transactions in analytics together. And doing so in a fashion that maybe humans aren't as involved. Maybe they're not involved at all. What do you mean by systems of intelligence, and how do your joint solutions address those? >> Yeah so, you know, one way to look at it is, I mean, so far if you look at how, sort of decisions are made and insights are gathered. It's we look at data, and between a combination of mostly, you know we try to get structured data, and then we try to draw inferences from it. And mostly it's human beings drawing the inferences. If you look at the promise of technologies like machine learning and deep learning. It is precisely that you can throw unstructured data where no patterns are obvious, and software will find patterns there in. And what we mean by systems of intelligence is imagine you're going through your business, and literally hundreds of terabytes of your transactional data is flowing through a system. The software will be able to come up with insights that would be very hard for human beings to otherwise kind of, you know infer, right? So that's one dimension, and it speaks to kind of the fact that there needs to be a more real time aspect to that sort of system. >> Is part of your strategy to drive specific solutions, I mean integrating certain IBM software on Power, or are you sort of stepping back and say, okay customers do whatever you want. Maybe you can talk about that. >> No we're very keen to take this up to a solution value level, right? We have architected our ISV strategy. We have architected our software strategy for this space, right? It is all around the cognitive workloads that we're focused on. But it's about not just being a platform and an infrastructure platform, it's about being able to bring that solution level above and target it. So when a client runs that workload they know this is the infrastructure they should put it on. >> What's the impact on the go to market then for that offering? >> So from a solutions level or when the-- >> Just how you know it's more complicated than the traditional, okay here is your platform for infrastructure. You know, what channel, maybe it's a question for Raja, but yeah. >> Yeah sure, so clearly, you know, the product will be sold by, you know, the community of Nutanix's channel partners as well as IBM's channels partners, right? So, and, you know, we'll both make the appropriate investments to make sure that the, you know, the daughter channel community is enabled around how they essentially talk about the value proposition of the solution in front of our joint customers. >> Alright we have to leave there, Stephanie, Raja, thanks so much for coming back in theCUBE. It's great to see you guys. >> Raja: Thank you. >> Stephanie: Great to see you both, thank you. >> Alright keep it right there everybody we'll be back with our next guest we're live from D.C. Nutanix dot next, be right back. (electronic music)