 Welcome to the special CUBE Conversation, here in the CUBE's Palo Alto Studios, I'm John Furrier, the co-host of the CUBE, and co-founder of SiliconANGLE Media, the CUBE. Here, Mike Was, who's the CEO and founder of Datometry, a hot startup, doing some interesting things in the cloud with data. We love data. Great, welcome to the CUBE Conversation. Thanks for coming in today. Thanks for having me. So you guys have an interesting approach, interesting love to model. You're in the tornado, as we say. You're kind of at the cloud game going on around you. A unique approach, very focused solution. I love the, I want to get into it because it's really compelling, but I want you to take a minute first to explain what the business is, Tom. How'd you guys found it? How'd you get here? What's the core value proposition that you guys do? All right. Well, Datometry is a venture-founded startup. We are still in the early stages, but we've seen great traction across the market. What we realized was really that as cloud catches on, everybody, every IT leader is scratching their heads, how do I get my data assets into the cloud? And one of the interesting elements there is that it's really the applications that's a difficult thing to move to the cloud because you end up rewriting them and very quickly people realize it's pretty bad ROI. And so they're looking for a solution that effectively allows them to take these applications, move them to the cloud as is, then cut down time be three, five years and obviously time and risk dramatically. And that exactly is Datometry. Think of it as VMware for databases, so to speak. So is it software as a service? Is it the software approach? What's the product-business model real quick? It's a software platform and we have a SaaS platform delivery as well. But think of it as really a, almost like a logical hypervisor that sits between the application and the database. So I buy software from you, I'm the customer. Yes, well you buy a subscription from us. Okay, great. So what's the core challenge that you guys saw? Because obviously the migrations to the cloud, I talked to Andy Jassy at Amazon and he's like, everyone's going to be in the public cloud. He talked to someone at VMware, hybrid cloud or multicloud. So again, different bias perspectives and I appreciate Andy's view but also customers want choice. But we know that everyone wants to go to the cloud. Every startup that I know never starts with the data center unless they have a unique situation, they go to the cloud. So cloud's a nice place to go. The challenge is how do they go there? And how do I just go to Microsoft and Amazon? How do I use multiple clouds? What's your solution do for that? So first off, we love all databases. We are multicloud. And what we really see IT leaders struggle with is kind of figuring out what's the right cloud for them. And that's very often driven by business decisions by all sorts of additional stuff that we kind of stay out of it. But then it very quickly comes down to the technical bits and pieces. And that is I have my applications today in the data center. I need to get them to the cloud and I don't want to spend these egregious amounts of money to rewrite everything. And if you think of it from a CIO perspective, this is an enormous amount of risk. Typically these migration projects are currently slated for three to five years which is pretty much all the time you got as a CIO. So this better work. Or you're going to look really, really bad at the end of that 10-year. You're going to be gone. You're going to spend a shelf life with a CIO could be. It's like a sports athlete, four years. Then you're a free agent again or you get renewed as a franchise player kind of thing. But this is interesting. I want to get into this. So moving the data is one thing. Migration to the cloud is the number one concern we hear from people because obviously there's some economic benefits. But there's all kinds of challenges around data, data mobility, lock-in. Am I going to get stuck in the cloud? I have a vendor selection on premise. I have multi-year licenses. It just becomes kind of like I give up. But they want a path there. So I want to ask you about a concept that you and I were talking about before we came on called the migration paradox. You brought that up. Explain what the migration paradox is for an enterprise who wants to move to the cloud. So it's a very interesting thing that when we start talking to people who are facing migration to the cloud, we realize that most of them actually have not really kind of fathomed yet how big of a challenge this is going to be. So a lot of people believe when they first look at database migration, they believe it's all about transferring the content. And that's kind of where their thinking almost ends. But the interesting or paradoxical thing is the hard thing about database migrations is actually not migrating the content of the database, but it's the applications. Moving the schema, moving the data, this is something that people have down. It takes three to six months. Amazon's got a great tool. You can snowmobile it, you can snowplow it, whatever they have, those tools, and they do it in a lot of oracle environments. We see that, but that's not enough. Exactly. So there's a lot of tooling that takes care of this, but then you're left with modifying or even rewriting your applications. And that typically is about 80 or even more percent of your actual cost. And so that's what we call the paradox about the migration, that you really need to take care of the applications. The content of the database, that's the easy part. So the paradox is interesting, but the other question that comes up is, which is good, thanks for clarifying that, is the other challenge is as a technical person, architect, CIO, whoever, where do I optimize my time? And this is a question that they're asking themselves. So I've got to ask you if I'm moving to the cloud, say now I got multi-cloud, I want to have Amazon and Azure, Microsoft Azure and Amazon Web Services. What do I spend my time? Do I optimize my database? Do I think about my database as the main source of truth, or do I think about databases in now a cloud world, which is borderless? You have applications that have database needs. So the kind of the thinking around the database becomes challenged in terms of how you think about it. And it changes fundamentally from what we've seen, let's say 10, 20 years ago, where the database is really kind of your ivory tower, where all the functionality, the complexity is really reciting there, and your applications are just satellites to it. And nobody ever wanted that. People rather solve data problems than database problems, but the database was always so dominant, and that was the only way of getting there. And so what we feel is going to be the model of the future is that you can actually focus on your applications. That's where your business is. This is where the value for your business comes from. And the database really becomes a commodity. When was the last time you bought a file system? These things just come with the platform, and you really want to focus on your investment in the application. The databases are in everything now, so it's not like the market for databases is basically every application. Exactly. Somebody asked you something, because I've been thinking about this, I want to get your thoughts on it, because you're a founder and CEO of a company, you're in the middle of the database world. We've been on the DevOps thing since day one, we've been so early, and back in the old days, the networks were, they dictated terms. If the network guys and the network architecture dictated what was enabled on top of it, in comes the cloud and DevOps, which smashes that paradigm. It says, hey, you know what? Let's make the network programmable. So that changes the game in terms of who's now in charge. DevOps creates an abstraction layer that allows for programmable infrastructures, infrastructure as code. We know that turned out that's the cloud. Hello, everyone's being disrupted. Now Cisco's of the world are actually doing better because they're actually bringing the programmability to a whole other level, so they're now right in that way. So at the end of the day, it was better for the networking industry to go DevOps. You could say the same thing's happening in the database world, where the database used to dictate terms. Now the applications are in charge. They're all going to have databases. So do you see a similar paradigm with that where I don't really care what database is out there, as long as my app runs. I should be programming my databases in the similar way that DevOps used to program, the network. So database as code? Absolutely. And you're touching on really the core revolution of IT over the last 10, 15 years, and that's virtualization. Everything in the IT stack has been shredded over the last 20 years. And the only thing that has been holding out is the database. And when you look closely, virtualization really means that two things that are otherwise bonded tightly together kind of get pried apart. An abstraction gets put in between. And suddenly you create that new geography for functionality for all sorts of things, including additional companies and opportunities that's kind of spring up in that space. We've seen this with software defined networking, as you said, with storage, with compute anyways. But databases were the last kind of corner in the data center that managed to hold out with this very strong, well, every application is bolted on my database principle. And that's exactly what we want to kind of take apart now, where we think the time is ready for this. And web services and services now, microservices and containers, you're seeing amazing new capabilities around software. Software needs databases. Sometimes it's stateless, stateful. These things are interesting new innovation areas. How is that changing the database world and the database architecture? And two, how is that obviously helping the cloud service providers? So this new paradigm that's happening at the application level is kind of flip things upside down, your thoughts and reaction to that. I think the database people have always been a bit envious of all the other adjacent spaces because they did have that virtualization component to it. And so you had a lot more flexibility. You have the liberty to just move your stuff, mix and match whatever makes both sense. And on the database side, you've always kind of been locked into, well, this is the choice that I made 10 years ago and it's going to be really difficult to get out of this. So we obviously believe that's the future. People really want that liberty. And what we see from talking to CIOs and IT leaders is that's one of the biggest problems that they've always had in the IT stack, that they just couldn't move the applications, that they were always kind of tied to decisions they made in the past to legacy. And we think it's time to kind of completely redefine that space. So you guys have been very successful. I'll just give you some props. I know that Amazon and Microsoft who are really killing it in the cloud. Obviously Amazon, really the president of creation, created cloud. Azure numbers are off the charts. Look at just Amazon stock, I mean, Microsoft stock price. It's been very successful. Satya Nutella's turned that ship to be completely cloud first and they really are moving a lot of the Microsoft legacy business into the cloud and with it their ecosystem, which is the enterprises. What has been some of the challenges that you've seen for those enterprises because we know that they have a relationship with Microsoft over many, many years on their side. Amazon has a cloud of cloud native enterprises and developers. Their challenge is to get to the cloud but they've got to really kind of build the airplane at 30,000 feet if you will. This is a huge challenge. The interesting little detail is that where we are right now with cloud, it's no longer for the early adopters. Ironically, it's actually not the early adopters. The early adopters are still busy kind of figuring their stuff out in their data center and doing blockchain and decentralized applications and doing ICOs really. And they want to be a software company and what have you. But it's actually the much more later or laterish adopters of technology that really go to the cloud now. And it's very interesting to see this because we kind of jump over the chasm if you will in the classic adopter curve. And what we see there is different verticals, different types of companies gravitate to different clouds simply because it matches their DNA. It matches their... What they have a track record in. I mean, Microsoft has years of experience. Even Oracle has success in the cloud. You've seen what they're doing. Well, they claim it, we'll see the numbers. But they still got the licensing issue as a database vendor. But I think the IT thing you brought up is interesting. I want to just double click on that because IT has changed. I think there's a realization now in IT that, okay, it's a serverless model. It's not about provisioning top of rack anymore and doing all this data center traditional stuff. It's about cloud operations on-prem and also on the cloud. How do you guys fit into that story? Because is that the kind of customer you guys wanted to reach? Or is it someone doing a full wholesale change over? What's the kind of profile customer? Because they're looking at the cloud, saying, hey, we get the IT game has changed. It's a services business. We got security challenges. We got this multi-cloud. Where do you fit it? So there is not kind of a cookie cutter customer for us, but it's really the Fortune 500, the Global 2000. It's the big guys. And what we see there is they're kind of trying to figure this out for them right now. And that's a fantastic challenge to kind of be part of. At the same time, they are the ones who realize that the only thing that they love about their database is the query interface and maybe the way how they get data in and out. What they really love about it is operations and maintenance. And so they look at the cloud databases now as, wow, somebody kind of reduced all of this to an API. And I don't need to have all this other stuff, all the operations, all the stuff that I always hated about my database is suddenly gone. And so for them, this is a monumental shift. Yeah. This is unique. They're like freed from the hostage situation that's been in place for many years. That's an interesting way of putting it. So they really like the idea that it's down to an API. And more importantly, they don't have to tune the heck out of this anymore. I've been in databases for 20 years and one of the big disciplines has always been tune this thing and kind of shave a couple of milliseconds here and there. And that was primarily because there was always a fixed hardware footprint that constrained people. Now in the cloud, you have a single slider and you say more or less. And it's a couple of dollars a month more and you kind of get yourself out of that pickle. And also you now have IT operations, another area that's been kind of automated away, not automated in a good way to see everything in terms of latency, what not. So I got to ask you the question around from a database perspective since you're a pro at it. The multi-cloud dilemma is interesting because there's no doubt that multi-cloud is on a lot of people's minds, especially ones have large applications because there's different use cases. I have a relationship with this or this is good for me, that's good for me. But also working workloads around clouds. I want to move to better price on Azure down the road. So there's definitely a leaning into the multi-cloud. What is the trade-offs for multi-cloud in your opinion? What's the pros and cons for someone who's thinking about migrating their applications and data to a multi-cloud architecture? I think we've seen only kind of the very beginning of the multi-cloud story yet. And when you look at the different clouds, kind of from a performance or a scale perspective, there are a lot of similarities, there are a couple of interesting differences. But one thing that from a database perspective is going to be really interesting is these databases that we see on the clouds that are really integrated now, they have access to special hardware, what have you. They're really fundamentally a part of the cloud and kind of really become commodity. And so for an enterprise, one of the things that they need to kind of go through is what's the premium for multi-cloud that they're willing to pay versus optimizing for a specific cloud? And it's not an easy decision. And what's the difference between those two? Well, as a separate cloud, as a third party database on a cloud, you're effectively kind of catering to the common denominator across all of us. You're just hosting on their cloud basically. Right. And so you take what you get and you're not really optimizing that well for a specific cloud. And that kind of gives you product completely different look and feel than let's say an integrated cloud database. And at the same time, it kind of gives you that freedom of being multi-cloud and flexible. And that's the thing where I think Julie's still out. You mean integrated side. You mean talking about the integrated side. Well, for the third party database vendors, where okay, you have that freedom to just move across the clouds. You pay a premium for that compared to the databases that are integrated. But it's unclear today, I think, is that premium really worth it? And how will people deal with this going forward? I mean, basically, it's a moving train because they have to constantly chase latency basically run operations on the cloud. Someone else's cloud basically. Exactly, yeah. So going integrated, it's to me, a better approach if I was going to look at that technically. So how do I deal with that? Because that's will be the preferred. Now, as a developer or an IT guy, well, integrated database on Microsoft, I got an integrated database on Amazon, two different code bases. They got, I got EC2, S3 and all the stuff on Amazon. And I got different, do I have to hire different coders to do that? So what's your response to that? We let you take whatever you've got and whatever your skill set is today and move that across different platforms, across different databases so that you can find out what's the one that's best for your business. And we love all the different coders for each platform or do that for me. You can keep whatever you have. Whatever skill set you already have in-house, whatever familiarity you have for the database that your folks have already kind of started out with, you just take that along to the next platform. And that's a very unique liberty find. They don't want to have to hire new people. The whole goal is to have people focus on high yield tasks. And they'll be productive on day one. They hit the ground running. All right, so what's the ideal customer for you? Obviously you have a great relationship with the cloud, sounds like on the integrate side. Well, if someone's watching this video or might be interested in your product, what are some of the pain points that they would have? What are some of the conversations that they may be having in their meetings that would tell them that they should be talking to you guys? Absolutely. So most of our customers, as we talked about before, really come through our partners. Our partners tend a lot to gain from this. And so we go to market with the big cloud service providers. And what we see as a typical pattern is people have kind of matured in their journey to the cloud from just being aspirational, I want to go to the cloud to, I figured out which cloud I want to go or even have kind of done some experimentation. And so at that point, they're ready to pull the trigger. Say, okay, I realized this is the cloud, this is the data warehouse that I want to go to. And then they realize the next step, okay, this would take me three to five years to migrate all that stuff. And it's gonna take me anywhere between five to 10 to $20 million. And it's this enormous amount of risk. And that is usually the point at where we have that conversation with that customer to kind of completely upend that equation. Say, well, we can do this in actually just a few months at a fraction of the cost at a fraction of the risk. Yeah, that could have cost them 20 million to move it. All right, here's two final questions for you to end this segment. Thanks for coming, it's been great to chat with you. First one is to the decision maker of your customer base. If you're on a joint call with one of the big clouds, what's the message to the decision maker? If you had to kind of like just tighten up that message. And the second part is to the user, the person who gets to move the action over to audiences, decision maker, what's the main benefits of what you guys do? Is it the person who's going to end up deploying it? What's the message to those guys? So for the decision maker, I think the message is simply you cannot afford not talking to us because the economics of what we do are just so overwhelming or not going with us is going to be really overwhelming for you. And so that is really about we let you get to the cloud quick. What we see there right now is it's not about kind of change your platform in your hardware. Smart enterprises, I think, get that huge opportunity right now to outmaneuver their competition if they get to the cloud rapidly. And so this is really at the decision maker's mind of all the major companies. For the end user, the message is simply keep what you're doing. Keep doing what you're doing right now. You don't need to kind of acquire a new skill set. Now you can, as you go to the cloud, we don't keep you on whatever you do right now. We give you the ability to mix and match to do all sorts of things. But you don't need to kind of just relearn everything just because you moved to the cloud. Awesome. Mike, thanks for coming and explaining value purposes and talking about some of the industry trends around the database. Certainly everyone knows I love to talk about databases, especially when you have the big guys out there. And some of them have been called hostage holders. But I mean, the freeing up of the data is what we believe in. We think that the cloud and movements like decentralized internet around decentralized applications is a big trend coming. Blockchain certainly significant when the ICO hype dies away in the next year or so, you're going to start to see a lot new, new generation of applications in charge and programming the data. So I think that's going to be big. It's a cute conversation here in Palo Alto. I'm John Furrier. Thanks for watching.