 From around the globe, it's theCUBE with digital coverage of DockerCon Live 2020 brought to you by Docker and its ecosystem partners. Hey, welcome back everyone to the DockerCon virtual conference DockerCon 20 being held digitally online. This is theCUBE's coverage. I'm John Furrier, host of theCUBE. This is theCUBE virtual, Cube Digital. We're getting all the remote interviews. We're here in our Palo Alto studio quarantined crew all getting the data for you. We've got Peter Quagenti, who's the chief marketing officer of Cockroach Labs, a company that we became familiar with last year. They had the first multi-cloud event in the history of the industry last year, notable milestone. Hey, first is always good, you're still around. So first you got the first position. Peter, great to see you. Thanks for coming on theCUBE for DockerCon 20. Thank you, John. Thanks for having me. So it's kind of interesting. I mentioned that tidbit to give you a little bit of love on the fact that you guys ran, who are part of the first multi-cloud conference in the industry. Okay, now that's all everyone's talking about. You guys saw this early. Take a minute to explain Cockroach Labs, why you saw this trend, why you guys took the initiative and took the risk to have the first ever multi-cloud conference last year. So that's news to me that we were the first actually. That's a bit of a surprise because for us, we see multi-cloud and hybrid cloud as the obvious. I think the credit really for this belongs with folks like Gartner and others who took the time to listen to their customer, right? Took the time to understand what was the need in the market, which what I hear when I talk to CIOs is cloud is a capability, not a place, right? They're looking at this and saying, yes, I have a go to cloud strategy, but I also have made massive investments in my data center. I believe I don't want to be locked in yet again to another vendor with proprietary APIs for proprietary systems, et cetera. So what I hear when I talk to customers is I want to be multi-cloud, show me how, show me how to do that in a way that isn't just buying from multiple vendors, right? Where I have cost arbitrage, show me a way where I can actually use the infrastructure in a creative way. And that really resonates with us and it resonates with us for a few reasons. First is we built a distributed SQL database for a reason, right? We believed that what you really need in the modern age for global applications is something that is truly diverse and distributed, right? You can have a database that behaves like a single database that lives in multiple locations around the world, but then you also have things like data locality. So you can say, okay, with German data, stays in Germany because of German law, you know, but when I write my application, I don't want to write to different, each of these things differently. Now, the other reason is customers are coming to us and saying, I want a single database that I can deploy in any of the cloud providers, you know? Azure SQL is a phenomenal product. Google Spanner is a phenomenal product, but once I do that, I'm locked in. Then all I have is theirs. But if I'm a large global auto manufacturer or if I'm a startup that's trying to enter multiple markets at the same time, you know, I don't want that. I want to be able to pick my infrastructure and deploy where I want and how I want. And increasingly, you know, we talk to the large banks and they're saying, I've spent tens or even hundreds of millions of dollars on data centers. I don't want to throw them out. I just want better utilization than the 15 to 20% that I get from deploying software on bare metal, right? I want to be able to containerize. I want to be able to, you know, cloudify my data center. And that have ultimately what we see more and more is what they call a tripod strategy where your own data center and two cloud providers behaving as a single unit for your most important applications. That's awesome. And I want to thank you for coming on too for DockerCon 20 because this is an interesting time where developers are going to be called to the table in a very aggressive way because of COVID-19 crisis is going to accelerate and certainly pull the future forward ahead of most people thought. I mean, we in the industry, we are inside the ropes, if you will. So we've been talking about, you know, stateless applications, stateful databases and all the architectural things that's got that longer horizon. But this is an interesting time because now companies are realizing from whether it's the shelter in place at scale problems that emerge to the fact that I got to have high availability at a whole nother level. This kind of exposes a major challenge and a major opportunity. We're expecting projects to be funded, some not to be funded, things to move around. I think it's going to really change the conversation as developers get called in and saying, I really got to look at my resources at scale. The database is a critical one because you want data to be part of this data plane, if you will, across clouds. What's your reaction to this? Do you agree with that? The future has been pulled forward and what's Cockroach doing to help developers manage this? Yeah, John, I think you're exactly right. And I think that is a story that I'm glad you're telling because I think there's a lot of signal that's happening right now but we're not really thinking about what the implications are. And we're seeing something that's, I think, quite remarkable. We're seeing within our existing customer base and the people we've been talking to feast or famine. And in some cases, feast and famine in the same company. And what does that really mean? You know, we've looked at these graphs for what's going to happen, for example, with online delivery services. And we've seen the growth rates and this is why they're all so valued, why Uber invested so big in Uber Eats and these other vendors. And we've seen these growth rates and saying, this is going to be amazing in the next 10 years we're going to have this adoption. That five to 10 years happened overnight, right? We were so desperate to hold on to the things that are what matter to us and the things that make us happy on any given day. You know, we're seeing that acceleration, like you said, it's all of that, the future got pulled forward, like you had said, that's remarkable. But were you prepared for it? Many people were absolutely not prepared for it, right? They were on a steady state growth plan. And we have been very lucky because we built an architecture that is truly distributed and dynamic. So, you know, scaling and adding more resilience to a database is something we all learned to do over the last 20 years as data intensive applications have mattered. But with a distributed SQL and things like containerization on the stateless side, we know we could just truly elastically scale, right? You need more support for the application out of something like Cockroach. You literally just add more nodes and we absorb it, right? Just like we did with containerization where you need more concurrency, you just add more containers. And thank goodness, right? Because I think those who were prepared for those things, we work with one of the large delivery services. Overnight, they saw a jump to what was their peak day at any point in time now happening every single day. And they were prepared for that because they already made these architectural decisions, right? If you weren't in that position, if you were still on legacy infrastructure, you're still trying to do this stuff manually where you're manually sharding databases and having to increase the compute under your model list, you were in trouble and you're feeling. You know, that's interesting, Peter. You bring that up and it reminds me of the time if you go back in history a little bit, just not too far back. I mean, I'm old enough to go back to the 80s and remember all the different inflection points. They all had their all key characteristics. I see computer revolution, TCP, IP. I mean, you pick your spots. There's always been that demarcation point or line in the sand where things change. But let's go back to around 2004 and then 2008. During that time, those legacy players out there kind of was sitting around sleeping at the switch and incomes open source, incomes Facebook, incomes roll your own. Hey, I'm going to just run, I'm going to run open source. I'm going to build my own database. And that was because there was nothing in the market. And most companies were buying from general purpose vendors because they didn't have to do all the due diligence. But the tech savvy folks could build their own and scale and that changed the game. That became the hyperscales in the rest of history. Fast forward to today, because what you're getting at is this new inflection point. There's going to be another tipping point of trajectory of knowledge, scale, that's completely different than what we saw just a year ago. What's your reaction to that? I think you're exactly right. What we saw, and I've been lucky enough, same like you, I've been involved in the web since the very early days. I started my career at the beginning. And what we saw with web 1.0 and the shift to web 2.0, web 2.0 would not have happened without open source. And I don't think we give them enough credit if it wasn't for the lamp stack, if it wasn't for Linux, if it wasn't for this wave of innovation, and it wasn't even necessarily about rolling around. Yeah, the Facebooks of the world could go hire their own engineers to go and improve my SQL to make it scale. That was of course a possibility, but the democratization of that software is where all of the success really came from. And I lived on both sides of it in my career as a both an app developer and then as a software executive in that window and got to see it from both sides and see the benefit. I think what we're entering now is yet another inflection point, like you said. We were already looking at it. I think the move from traditional applications with simple logic and simple rules to now highly data intensive applications where data is driving the experience, models are driving the experience. I think we were already at a point where ML and AI and data intensive decision making was going to make us rewrite every application we had and that needed a new infrastructure. But I think this is going to really force the issue. And it's going to force the issue at two levels. First is the people who were already innovating in each of these industries and categories were already doing this. They were already cloud native. They were already built on top of very modern third generation databases, third generation programming languages, doing really interesting things with machine learning. So they were already out innovating but now they have a bigger audience. And then if you're a traditional and all of a sudden your business is under duress because substantial changes in what is happening in the market used to still retailers still had strength with footprint as of last year. We don't, we think about e-commerce versus traditional retail. Yeah, it was on a slow decline. There were lots of problems but there was still a strength there. That happened changed overnight, right? Now a lot of the new sources dried up. So what are you going to do and how are you going to act? If you've built your entire business, for example, on legacy databases from folks like Oracle and old monolithic ways of building applications, you're simply not adaptable enough to move with changing times. You're going to have to start, we used to talk about every company needed to be a software company. That mostly happened but they weren't all very good software companies. I would argue that the next generation is you have to be a great software company and great data scientists. Look at the software companies that have risen to prominence in the last five to 10 years. Folks like Facebook, folks like Google, folks like Uber, folks like Netflix. They use data better than anyone else in their category. So they have this amazing app experience and leverage data and innovate in such a way that allow them to just dominate their category. And I think that is going to be the change we see over the next 10 years. And we'll see who exits what is obviously going to be a downturn. We'll see who exits on top. Well, it's interesting to have you on. I love the perspective and the insights. I think that's great for the folks out there who haven't seen those ways before. Again, this way is coming. Let's go back to the top when we were talking about what's in it for the developer because I believe there's going to be not a renaissance because it's always been great, but the developers even more are going to be called to the front lines for solutions. I mean, these are first generation scale problems that are going to be in this whole next generation modern error that's upon us. What are some of the things that's going to be that lamp stack like experience? What are some of the things that you see? Because you guys are kind of a tell sign in my opinion, cockroach, because you're thinking about things in a different construct. You're thinking about multi-cloud, you're thinking about state, which is the database challenge. Stateless has kind of been around restful APIs, stateless data, service meshes. Kubernetes is also showing a cloud native and the services microservices or service orientation is the future. There's no debate on that. I think that's done. Okay, so now I'm a developer. What the hell am I going to be dealing with for the next five years? What's your thoughts? Well, I think the developer knows what they're already facing from an app perspective. I think you see the rapid evolution in languages and in deployment and all those things are super obvious. You just need to go and sit in on what I'm sure that all the DockerCon sessions to see what the change to deployment looks like. I think there are a few other key trends that developers should start paying attention to that are really critical. The first one and only loosely related to us is ML Ops. I think what you're, just like we saw with Dev and Ops, starting to come together so we can actually develop and deploy in a super fast, iterative manner, the same thing's now going to start happening with data and all of the work that we do around deploying models. And I think that that's going to be a pretty massive change. You think about the rise of tools like TensorFlow, you know, some of the developments that have happened inside of the pod providers. You know, I think you're seeing a lot there as a developer. You have to start thinking as much like a data scientist and a data engineer as simply somebody writing front and code, right? And I think that's a critical skill that the best developers are already building. I think we'll continue. I think then the data layer has become as important or more important than any other layer in the stack because of this. And you think about, once again, how the leaders are using data and the interesting things that they're doing, the tools you use matter, right? If you are spending a lot of your time trying to figure out how to shard something and how to make it scale and how to make it durable when instead you should be focused on just the pure capability, that's a ridiculous use of your time, right? That is not a good use of your time. We're still using 20 to 25 year old open source databases for many of these applications when they gave up their value probably 10 years ago, honestly, you know, we keep wallpaper over it, but it's not a great solution. And unfortunately, no SQL while it fixed some of the issues with scale and elasticity, it's like you and I starting a business and saying, okay, everyone speaks English, but because we're global, everyone's going to learn Esperanto, right? That doesn't work, right? So it works for a developer, but if you're trying to do something where everyone can interact, this is why this entire new third generation of new SQL databases have risen. We took the distributed architecture from the SQL. Hold on for a second. Can you explain what that means? Cause that's a, I think a key topic. I want to just call that out. What is this third generation database mean? Sorry, I speak about it like everyone sees it. I think it's super important. It's just a highlight. Just take a minute to explain it. We can get into it. There is an entire new wave of database infrastructure that has risen in the last five years and it started actually with Google. So it started with Google Spanner. So Google first was the first to face most of these problems, right? They were the first to face web scale, at least of the scale we now know it. They were the first to really understand the complexity of working with data. They had their own no SQL. They had their own way of doing things internally and they realized it wasn't working. And what they really needed was a relational database that spoke traditional ANSI SQL, but scaled like their no SQL counterparts. And there was a white paper that was released that was the birth of Spanner. Spanner was an internal product for many, many years. They released the thinking into the wild and then it just started this wave of innovation. That's where our company came from and there were others like us who said, you're right. Like let's go build something that behaves like we expect a database to behave with structure and this relational model. And anyone can write SQL, they can use it. It's the simplest API for most people with data, but it behaves like all the best distributed software that we've been using. And so that's how we were born. Our company was founded by ex Googlers who had lived in this space and decided to go and scratch the itch, right? And instead of doing a product that would be locked into a single cloud provider, a database that could be open source, it could be deployed anywhere, it could cross actual cloud providers without any hiccups, right? And that's been the movement. And it's not just us, there were other vendors in this space and we're all focused on really trying to take the best of the both worlds that came before us. The traditional relational structure, the consistency and asset compliance that we all loved from tools like Oracle frankly, right? And Microsoft who we really enjoyed. But then the developer friendly nature and the simple elastic scalability of distributed software. And that's what we're all seeing that our company for example has only been selling a product from the last two years. We found in five years ago, took us three years just to write to the software that we would be happy selling to a customer where on what we believe is probably a 10 to 15 year product journey to really go and replace things like Oracle. But we started selling the product two years ago and we saw 300% growth year over year. We're probably one of the fastest growing software companies in America, right? And it's all because of the latent demand for this kind of a tool. Yeah, that's a great point. I'm a big fan of this third wave. Could I see it? I mean, if you look at just the macro tailwinds in the industry, billions of edge devices, immersion of all kinds of software. So that means you can't have one database. I always said this when I interviewed Andy Jassy and others. You can't have one database. You can't just, it's physically impossible. You need data. And whatever database fits the scene, whatever. You want to have data being stored, but you got to have it real time. You got to have actionable. You have to have software intelligence into how to manage the data. So I think the data control plane or that layer, I think it's the next interoperability wave. Because without data, nothing really works. Machine learning doesn't really work well. You want the most data. I think cybersecurity is a great early use case because they have to leverage data fast. And so you start to see some interesting financial services, cyber. What's your thoughts on this? Can you share from the cockroach labs perspective from your database? You got a cloud. What are some of the adoption use cases? Who are those leaders? What is, you can name names if you have them, if not name the use case. What's the cockroach approach? Who's winning with it? What's the, what's it look like? Yeah, that's a great question. And you nailed it, right? The data volumes are so large and they're so globally distributed. And then when you start layering in the data streaming in from devices that then have to be weighed against all of these things. You want a single database, but you need one that will behave in a way that's going to support all of that and actually is going to live at the edge, like you're saying. And that's where we have been shining. And so our use cases are, and unfortunately I can't name many names, but for example in retail, we're seeing retailers who have that elasticity and that scale challenge with commerce. And what they're using us for is then we're in all of the locations where they do business. And so we're able to have data locality associated with the businesses and the purchases in those countries. And however, we only have single apps that actually bridge across all of those environments. And with a distributed nature, we were able to scale up and scale down truly elastically because we spread out the data across the nodes automatically. And what we see there is, retailers do have up and down moments. And you talk about people who leverage the financial structure of the cloud in a really thoughtful way. Retail is a shining example of that. I remember having customers that had 64 times the amount of traffic on Cyber Monday that they had on the average day. In the old data center world, that's what you bought from. That was horrendous. In a Kolo environment, it's still horrendous. Even in public cloud providers, if you're having to go and change your app to ramp every time, that's a problem. With something like a distributed database and with containerization, you could scale much more quickly and scale down much more quickly. That's a big one for us. Streaming media is another one. Same thing with data locality in each of these countries. You think about it, it's something like Netflix or Hulu. They have shows that are unique to specific countries. They then have that user behavior, all that user data. They want data sovereignty. What you watch on Netflix is some very rich personal data. And we all know how that metadata has been used against people before. So it's no surprise that you now have countries saying, no, there's going to be regulation around where that data can live and how. And so once again, it's something like cockroach where you can have that global distribution but data locality, where we can lock data to certain nodes in certain locations. That's a big one. There's no doubt in my mind. I think there's such a big topic. We probably do more interviews just on the COVID-19 data problem that they have. The impact of getting this right is a nerd problem today, but it is a technology solution for society globally in the future. Zero doubt in my mind on that. So Peter, I wanted you to get the last word and to give a plug in to the developers that are watching out there about Cockroach. Why should they engage with you guys? What can you offer? Is there anything you want to share about the company to the audience here at DockerCon 2020? Take us home. Thank you, John. I'll keep the sales pitch to a minimum. I'm a former developer myself. I don't like being sold. So I appreciate it. But we believe we're building what is the right database for the coming wave of cognitive applications. And specifically we've built what we believe is the ideal database for distributed applications and for containerized applications. So I would strongly encourage you to try it. It is open source. It is truly cloud native. We have free education, so you can try it yourself. And once you get into it, it is traditional SQL that behaves like Postgres and other tools that you've already known and loved. And so it should be very familiar. If you've come up through any of these other spaces, it'll be very natural, it's Postgres compatible, integrates to the number of ORM, so as a developer it'll just plug right into the tools you use and we're on a rapid journey. Like we believe we can replace that first generation of technology built by the oracles of the world and we're committed to doing it. We're committed to spending the next five to 10 years in hard engineering to build that most powerful database to solve this problem. Well, thanks for coming on and sharing your awesome insight and historical perspective. You've got a lot of experience. We believe and we want to share the audience that in this time of crisis, more than ever, the focus on critical nature of operations because coming out of this, it is going to be a whole new reality and I think the best tech will win the day and people will be building new things to grow, whether it's for profit or for societal benefit. The impact of what we do in the next year or two will determine a big trajectory and new technology, new approaches that are dealing with the realities of infrastructure, scale, working at home, sheltering in place, to coming back to a hybrid world. I mean, we're coming virtualized, Peter. We've been virtualized, the media, the lifestyle, not just virtualization in the networking sense but fun times, it was going to be challenging. So thanks for coming on. Thank you very much, John. Okay, we're here for DockerCon 20, virtual conferences at CUBE, virtual segment. I just want to thanks you for watching. Stay with us, we've got stream all day today and check out the sessions. Jump in, it's going to be on demand. There's a lot of videos it's going to live on and thanks for watching and stay with us for more coverage and analysis. Here at DockerCon 20, I'm John Furrier. Thanks for watching. From the CUBE studios in Palo Alto in Boston, connecting with thought leaders all around the world, this is a CUBE conversation.