 Live from Miami, Florida, it's theCUBE. Covering IBM's data and AI forums. Brought to you by IBM. We're back in Miami. You're watching theCUBE, the leader in live tech coverage and we're covering the IBM data and AI forum in the port of Miami. Mathias Funka is here. He's the director of offering management for hybrid data management, everything data. Mathias, great to see you. It's great to have you. It's great to be here with you. We're going to talk database. We're going to talk data warehouse, everything data. You know, the database market, you know, 10 years ago, 12 years, it was kind of boring, right? And now it's like data is everywhere. Database is exploding. What's your point of view on what's going on in the marketplace? You know, I mean, it's funny to use the word boring because I think it's the boring stuff that really matters. Nowadays, to get things to value, get people to value with the solutions you want to build, the modernizations they are seeking to do on the data estates, the challenges they have in embracing multi-cloud data architectures. So to get to value, you have to take care of the boring stuff. How real is multi-cloud? I mean, I know multi-cloud is real and that everybody has multiple clouds, but is multi-cloud a strategy or is it sort of a symptom of multi-vendor and it's just we kind of ended up here with shadow IT and everything else? I think it's a reality. And yes, it should be a strategy, but I think more clients than not, they define themselves being exposed to these as a reality with different line of businesses, acquiring data estates, running on different locations, different clouds, you know, and then companies are challenged if you want to bring it all together and actuate the value of that data and make it available for analytics or AI solutions, you know, you got to have a strategy. So IBM is one of the few companies that has both a cloud and an aggressive multi-cloud strategy. You know, Amazon's got outposts a little bit here. Microsoft, I guess, has some stuff, but generally speaking, Oracle's got a little bit here, but IBM has both a cloud, so you'd love people to come into your cloud, but you recognize not everybody's going to come into your cloud, so you have an aggressive multi-cloud strategy. Why is that? What's the underpinning of that strategy? Is it openness? Is it just market, you know, total available market? Why? So, first of all, yes, we have a strong portfolio on IBM Cloud and we think, you know, it's the best in terms of, you know, integration with other cloud services, the performance you get on the different data services, but we also have a strategy that says we want to be where our clients want to go and many clients might have committed already on a strategic level to a different cloud, whether that's AWS, you know, or IBM Cloud, or Asia, and so we want to be where these clients want to go and our commitment is to really offer them, you know, a complete portfolio of data services that support different workloads and a complete portfolio in terms of, you know, the IBM proprietary set of technologies as well as open source technologies. Give clients choice, but then make them available across that universe of multi-cloud, on-premise, in a way that they get a consistent experience. And, you know, I mean, you are familiar with the term divide and conquer, right? I like to talk about it as, you know, unified to conquer. So, our mission is really unified experience and unified the access to different capabilities available across that multi-cloud architecture. So, is that really the brand promise? Going to unify across clouds? Absolutely, that's our mission. And that, what's the experience like today and what is sort of the optimal outcome that you guys are looking for? Being able to run any database on any cloud, anywhere, describe that. So, I think we talk about chapter one and two of the cloud, right? When it comes to chapter one, in my view, chapter one was very much about attracting people to the cloud by offering them a set of managed services that take away the management headaches and, you know, the infrastructure management aspects. But when you think about chapter two, when you think about how to run mission critical workloads on a cloud or on premise, you know, you want to have the ability to integrate data states that run in different environments. And we think that OpenShift is leveling the playing field by avoiding lock-in, by giving clients the ability to basically abstract from those proprietary cloud infrastructure services and mechanisms, and that gives them freedom of action. They can deploy a certain workload in one place and then decide six months later that they are better of moving that workload somewhere else. Yes, so OpenShift is the lynch pin to that cross-cloud integration, is that right? Correct, and with the advent of the rise of the operator, I think you see the industry closing the gap between the value proposition of a fully managed service and what a client-managed OpenShift-based environment can deliver in terms of automation, simplicity, and value. Let's talk about the database market and kind of what's happening. You got transactional database, you got analytic database, you got legacy data warehouses, you got new emerging databases that are handling unstructured data, you got NoSQL, not only SQL, lay out the landscape and what's IBM's strategy in the database space? So our strategy has, so starting with the DB2 family, right? We have introduced about two years ago, we introduced something called the CommonSQL Engine that gives you a consistent experience from an application and user perspective in the way you consume data for different workload types, whether that's transactional data, analytical use cases, big data overdue, or fast data solution event-driven data architectures. Everything with a consistent experience from a management perspective, from a working behavior perspective in the way you interact with this as an application. And not only that, but also then make that available on-premises, in the cloud, fully managed, or now OpenShift-based on any cloud, right? So our, I would say, our commitment right now is very much focusing on leveraging OpenShift, leveraging CloudPack for data, as a platform to deliver all these capabilities, DB2 and OpenSource in a unified and consistent, I would say, with a unified and consistent experience on anybody's cloud. It's like Watson anywhere it was first, you know, like six months ago when we announced it, and I think now for us doing the same with data and making then data, make it easy for people to access data wherever it resides is really key. But Tias, what's IBM's point of view on the degree of integration that you have to have in that stack from hardware and software? So some people would argue, well, you have to have the same control plane, same data plane, same hardware, same software, same database on-prem as you have in the cloud. What's your thoughts on that degree of homogeneity that's required to succeed? So I think it's certainly something that companies strive to get to simplify the data architectures, unify, consolidate, reduce the number of data sources that you have to deal with. But the reality is that the average enterprise client has 168 different data services they have to incorporate. So to me, it's a moving target and while you want to consolidate, you will never fully get there. So I think our approach is we want to give the client a choice, different choice in terms of technologies for the same workload type, potentially, whether it's a post-gres for transactional workloads or a DB2 for transactional workloads, whatever fits the bill, right? And then at the same time abstract or unify on top of that by when you think about operators and open shift, for instance, we invest in operators leveraging a consistent framework that basically provides a homogeneous set of interfaces by which people could deploy and lifecycle manage a post-gres instance or DB2 instance. So you need only one skill set to manage all these different data services and it reduces total cost of ownership, it provides more agility and accelerates time to value for this client. So you're saying that IBM strategy recognizes the heterogeneity within the client base. You're not taking, even though you might have a box somewhere in the portfolio, but you're not taking a, you need this box only strategy, the God box, this is the hammer and every opportunity is a nail. Yeah, we are way beyond that. So we are much more open in the way we embrace open source and we bring open source technologies to our enterprise clients and we invest in the integration of these different technologies so the value of those can be actuated in a much more straightforward fashion. The thing about CloudBack for data and the ability to access data that resides in different open source, different repositories, IBM and third party, but then make that data accessible through data virtualization or through governance, applying governance to the data so that a data scientist can actually retrieve that data for his work, that is really important. Can you argue that that's less lock in than say like they say the God box approach or the cloud only approach? Yeah, absolutely. How so? How so because, well, because we give you choice to begin with, right? And it's not only choice in terms of the data services and the different technologies that are available, but also in terms of the location where you deploy these data services and how you run them. Okay, so to me it's all about exit strategies. If I go down a path and a path doesn't work for me, how do I get out? Exactly. Is that a concern of customers in terms of risk management? Yeah, I think look, every customer out there, I dare say has a data strategy and every customer needs to make some decisions. But you know, there's only so much information you have today to make that decision. But as you learn more, your decision might change six months down the road. And you know, how to preserve that agility as a business to cause corrections I think is really important. So okay, hypothetical, but this happens every day. You got a big portfolio companies, they've done a lot of M&A, they've got 10 different databases that they're running, they've got different clouds that they're using, they've got different development teams that are using different tooling, certainly different physical infrastructure. And they really haven't had a strategy to bring it all together. You're hired as the data architect or the CTO of the company and say, Matias, the CEO says, fix this problem. We're not taking advantage and leveraging our data. Where do you start? So of course, being IBM, I would recommend to start with Gladbeck for data as the number one data platform out there because eventually every company will want to capitalize on the value that the data represents. It's not just about the data layer, it's not just about the database. It's about the integrated solutions tech that gets people to do analytics over the data, derive insights from the data. That's number one. Even if it's not the IBM stack, I would always recommend to the client to think about a strategy that allows for the flexibility to change course and move workloads from one location to another or move data from one technology stack to another. And I think that kind of agility and flexibility and translates into risk mitigation strategies that every client should think about. So cloud pack for data, you say, okay, let's start there. I'm going to install that or I'm going to access that into the cloud and then what do I have to do as a customer to take advantage of that? Do I just have to point it at my data stores? What are the prerequisites? Yeah, so let's say you deploy that on IBM cloud, right? Then you have, you usually are invested already, so you have large data states either residing on service or in the cloud. You can pull those data sets in remotely without really moving the workload or the data sets into a cloud pack for data management environment by using technologies like data virtualization, right? Or using technologies like data stage and ETL capabilities to access the data. But you can also, as you modernize and you build your next generation application, you can do that within that cloud pack for data management environment with OpenShift. And that's what most people want to do. They want to do a digital transformation. They want to modernize the workloads, but they want to leverage the existing investments that they have been making over the last decade. Okay, so but there's a discovery phase, right? When you bring in cloud pack for data, you say, okay, what do I have? Go find it. And then it's bringing in necessary tooling on the development side with things like OpenShift. And then what? It magically virtualizes my data, is that? So just on that point, I think what matters much more going forward for these clients is how they can incorporate different data sets in OpenShift technologies or DB2 or other third party vendors that I don't want to mention right now, right? But what matters more is, so how do I make data accessible? How do I discover the data sets in a way that I can automatically generate metadata? So I have a business glossary, I have metadata and I understand where these data sets live. That's their business objective, business technology objective. For us to be able to do that and to watch knowledge catalog, which is part of Cloud Pack for Data, is a core component there, helps you with that. Auto discovery, metadata generation, basically generating or creating data sets in a way that they are now visible to the data scientists and the ultimate end user. What really matters, and I think what is our vision overall, is the ability to service the ultimate end user whether it's a developer, a data scientist, or a business analyst, so that they can get a job done without depending on IT. Yeah, so that metadata catalog is part of the secret sauce that allows the system to know what data lives where, how to get to it, and how to join it. It's one of the core elements of that integrated platform and solution stack. But what I think is really key here is the effort we spend in integrating these different components so that it works seamlessly, it is happening in an automated fashion as much as possible, and it delivers on that promise of a self-service experience for that person that sits at the very end of that chain. Great. Matthias, thanks so much for explaining that. Thank you Dave. Thanks for coming on theCUBE. Great to meet you. All right, keep it right there everybody. We'll be back with our next guest right after this short break. You're watching theCUBE from the IBM Data and AI Forum in Miami. We'll be right back.