 Live from San Francisco, it's theCUBE. Covering IBM Think 2019, brought to you by IBM. Welcome back to theCUBE. We are live at IBM Think 2019. I'm Lisa Martin with Dave Vellante. We're in San Francisco this year at the newly re-judged Moscone Center. Welcoming to theCUBE for the first time, Yosef de Vries, Director of IBM Cloud Databases. Yosef, it's great to have you on the program. Thank you very much. Great to be here. So, as we were talking before we went live, this is, I was asking you what you're excited about for this year's IBM Think. Only the second annual IBM Think, this big merger of a number of shows, kind of day minus one, kind of T minus one, everything really kicks off tomorrow. Talk to us about some of the things that you're working on. You've been at IBM for a long time, but cloud managed databases. Let's talk value there for the customers. Yeah, definitely. Cloud managed databases really, at its core is about simplifying adoption of cloud provided services and reducing the capital expense that comes along with developing your application. So, fundamentally, what we are trying to do is abstract the overhead that is associated with running your own systems. Whether it's the infrastructure management, whether it's the network management, whether it's the configuration and deployment of your databases, our collection of services really is about streamlining time to value of accessing and building against your databases. So, what we're really focused on is allowing the developer to focus on their business critical applications, their objectives, and really what they're paid for. They're paid to build applications. They're not paid to maintain systems. When we talk about the CIO office, the CTO office, they're looking at costs, they're looking at ways to reduce overall expenditures. And what we're able to provide with cloud managed databases is the ability not to have to staff an IT team, not to have to maintain and pay for infrastructure, not have to procure licenses. What have you? Everything that goes into standing up and managing those systems yourself, we provide that and we provide it on a consumption-based method. So, you basically pay for what you use and we have various ways in which you can interact with your databases and the charges that are associated with that. But it really is, again, about alleviating all of that overhead and that expense that is associated with running systems yourself. 15 years ago, when you went back to, before you started with IBM, it was obviously had IBM DB2, it had Oracle, SQL Server. I guess my SQL was around back then, that the LAMP stack was building out the internet, but databases were pretty boring back then. And then all of a sudden it exploded and the NoSQL movement happened in a huge way. It coincided with the big data movement. What happened? I think as we saw the space of this technology evolve and the variety of different kind of use cases cropping up, the development community kind of respond to that. And really what we try to do with our portfolio is provide that variety of database technology solutions to meet on any number of different use cases. And we kind of like to think about it broken down into two categories, your primary data stores. This is where your applications are writing and reading the data that has been stored. And then particularly to your point, this is what we call the auxiliary data services, for example, these are your in-memory caches, your message brokers, your search indexes, what have you. There's a plethora of different database technologies out there today that plug into any number of different use cases and application developers is attempting to fulfill. And more often than not, they're using more than one database at a time. And really what we're trying to do at IBM with our cloud managed database offering is provide a variety of those data services and database technologies to meet a variety of those use cases, whether it's mixing and matching or different kind of application workloads or what have you. We like to provide our customers with the choices that are out there today and the community at large. So many choices, am I hearing that it's kind of horses for courses? I mean, you got things like, like even niche is like, you know, a cumulo with fine grain, you know, security, a couch base, obviously, this one scales and then this one's easy to use, you take Mongo for text, really easy to use. Sort of different specialized use cases. How do you sort of squint through and how does IBM sort of match the right characteristics with the right technology? It's really, it's too prong. It's about understanding your user base, understanding and listening to your customers and, you know, really internalizing what are the use cases that they're looking to fulfill. It's also being in tune with the database technology in the market today. It's understanding where there are trends, understanding where there are new use cases cropping up and it's about building a deep enough engineering and operations team where we can quickly spin up these new offerings and again, provide that technology to our end customers. And it's about working with our customers as well and understanding their use cases and then sometimes making recommendations on what database technology or combination of databases would be best suited for their objectives. And I'm curious, one of the things that you mentioned in terms of what the developer's day-to-day job should be, is this almost IBM's approach to sort of aligning with the developer role and enabling it in kind of new ways? It is really about, I think, having sympathy and delivering on solutions in regards to that sympathy for the pains that they'd otherwise endured 10, 15 years ago when the notion of cloud manage anything really wasn't a thing yet, or it was just starting to emerge. And IBM and House has run their own systems for years and years, obviously, and the folks on my team that have come from other companies, they know that the pain, what pain is involved in trying to run services. So like I said, it's a little bit out of sympathy and it's a bit out of knowing what your users need in a cloud managed service, whether again, it's security or availability or redundancy, you name it. It's about kind of coming around to the other side of the table and say, I sat where you once sat and we know what you need out of your data services, so trust in us to provide that for you. How are the requirements different? I mean, things like recovery and resiliency, do I need asset compliance in this new world? And it's funny, that's a good question in that. We don't necessarily deal so much with database-specific requirements. Again, as I mentioned, we try to provide a variety of different database technologies and by and large, the users are going to know what they need, what combinations that they will need and we'll work with them if they're navigating their way through it. Really where we see more of the requirements these days are around the management characteristics, as you cited. Are they highly available? Are they backed up? What's your disaster recovery policy? What security policies do you have in place? What compliance, so on and so forth? It's really about presenting the overall package of that managed solution, not so much whether the database is going to be highly available versus consistent replication and what have you. I mean, that's in there and that's part of what we engage with our customers about. But also where we like to put a lot of emphasis is on providing those recognized database technologies so that there's a community behind them, there's an opportunity for the users to understand what it is that they need beyond just what we can sell them. It's really about selling the value proposition of, again, the management characteristics of the services. So who do you see as the competition? I mean, obviously the other big, the two big cloud providers, AWS and Azure, you're competing with them for the quality of offerings. Talk about how you fit. And Google's another one, or Oracle's another emerging one, even Alibaba is catching up quite a bit. And it really feels like a neck-to-neck race day after day that the way we try to approach our portfolio is focusing on deep, broad, and secure. Deep mean that there are a core set of database technologies where we're building the database itself, DB2, Cloudant, which is based off of CouchDV, and then broad, again, as I've been mentioning, having a variety of different database technologies, and then secure across the board, whether it's secure in how we run those systems, secure in how we certify them through external compliance certifications, or secure in how we integrate with security-based tooling that our users can take advantage of. And regarding our competitors, it really is, you know, one week it may be a new big data-at-scale type of database technology. Another day it might be, or another week, it might be deeper integrations into the platform. It might be new open-source database technologies. It might be a new proprietary database technology, but we're, you know, it's a constant, like I say, race to who can have the most robust portfolio. Developers like teenagers, they're fickle. Yeah, that too, that too. And we've got to be quick and, you know, you know how to respond to those demands. In this age of hybrid multi-cloud, where the average company has five-plus private cloud, public cloud, through inertia, through acquisition, et cetera, where is IBM's advantage there as companies are, I think we heard a stat the other day, Dave, that in 2018, 80% of companies migrated data and apps from public cloud. So in terms of this reality that companies live in this multi-cloud, where is IBM's advantage there? And where does your approach to cloud management services really differentiate IBM's capabilities? Really, there's, for the last couple of years, a tremendous amount of investment on building on the Kubernetes open-source platform. And even in particular to our cloud-managed database services, we have been developing and have been recently releasing a number of different databases that run on a platform that we've developed against Kubernetes. It's a platform that allows us to orchestrate deployments, you know, deletions of databases, backups, high availability, platform-level integrations, all a number of different things. And what that has allowed us to do when concerning a hybrid-type strategy is it makes our platform more portable. So Kubernetes is something that can run in the cloud, it can run in a private cloud, it can run on-premise, and this platform we're developing is something that can be deployed, which we do today for public cloud consumption, which can also be packaged up and deployed into a private cloud-type environment, and ultimately is portable in its leveraging of the Kubernetes technology itself. So we're not hamstringing ourselves to purely public cloud-type services, or only private cloud-type services. We want to have something that is abstracted enough that, again, it can move around to these different kind of environments. How important is open-source and how important is it for you to commit to the different open-source projects? I mean, there's so many, and you've got no limited resources, so how do you do that? Open-source is really critical, both in what we're building and what we're also offering. As we've talked about our users out there, they know what they often want, or sometimes we nudge them to the right or to the left, but generally speaking, it's around all the open-source technologies, and whatever may be trending for that current month is oftentimes what we're getting requested for, so it could be Postgres, it could be RabbitMQ, it could be Elasticsearch, what have you, and really we put a lot of emphasis on embracing the open-source community, providing those database technologies to our customers, and then it allows our customers to kind of benefit from the community at large too. We don't become, again, the sole provider of education and information about that technology. We are able to expose the whole community to our customers, and they're able to take advantage of that. A lot of, I hear a lot of complaints sometimes, particularly from folks that might list themselves in a marketplace for one cloud or another, that they feel like the primary cloud vendor might be nudging the customer into their proprietary database. What's IBM's position on that? Is that fair? Is that overblown? You know, we obviously have proprietary tech, particularly with DB2, and that's something we're going to continue investing and it's what we view as one of our kind of strategic top priority database technologies. We're very active developers in the couch community as well. I wouldn't consider that proprietary. But again, back to the point of- CouchDB, you know, as the steward of CouchDB. Exactly, that's right. Right, exactly. But again, firm believers in open source, and we want to give those opportunities to our customers to avoid those vendor lock-in type situations. We actually have quite a lot of interest from our EU customer base. And by and large, EU policies around antitrust and what have you, they tend to gravitate towards the open source technology because they know it's again portable. They can be using Postgres by IBM one month and if they no longer are satisfied with that, they can take their Postgres workloads and move them to another cloud provider. Ideally, they're coming from the other cloud providers to IBM. Well, I should be more specific. I mean, just in fairness, Dynamo is often cited. I suppose Google Spanner, although that's sort of a more of a niche, specialized database. If I understand it correctly, DB2 is, that's a hardcore transaction system. So you're not going to confuse that with, I don't think anyway, CouchDB. Although who knows, maybe there are some use cases there. But it sounds like you're not sort of nudging them to your proprietary, certainly DB2 proprietary. And CouchDB is one of many options that you offer. Certainly DB2 is one of our core products for our database portfolio. And we do want to push our customers to DB2. If it makes sense. Exactly, where it makes sense. And where there's demand for it. And if it doesn't make sense or there's not demand, we'll offer up any number of the other databases that we also offer. Excellent, you just have last question. As IBM Think, the second annual, kicks off really tomorrow for this developer audience that you were talking about a lot in our conversation. What are some of the exciting things that they're going to hear? Any sort of, obviously not breaking news, but where would you advise the developer community who's attending IBM Think, to go to learn more about how to manage databases and how they can really become far more efficient, do their jobs better? You know, databases are hard, plain and simple. They're particularly hard to run. And developers who are not necessarily, you know, database admins that are not database operators that they want to focus on building applications, are going to want to find solutions that alleviate that overhead of running the systems themselves. So, you know, to your question, we've got sessions all throughout the week where we're talking about our cloud offering and the future of where we're going with that. We've got a couple different sessions around our IBM Cloud Database portfolio. This is a lot of the open source database technologies we're running. We have demos in the Solutions Center and DB2 is flooded all around the conference as well. So, there's lots of different sessions focused on talking to the value proposition of IBM's Cloud Managed Database Portfolio across the board. A lot of opportunity for learning. Well, Yosem Defri, thank you so much for joining Dave and me on theCUBE this afternoon. Thank you very much, it was great. And for Dave Vellante, I'm Lisa Martin. You're watching theCUBE live from IBM Think 2019. Day one, stick around, we'll be right back with our next guest.