 Okay, welcome back to SuperCloud 2 live here in Palo Alto. I'm John Furrier with Dave Vellante. Again, all day, wall-to-wall coverage. Just had a great interview with Walmart. We've got a next interview coming up. You're going to hear from Bob Mugley, interested in handing two experts, both experienced entrepreneurs, executives in technology. We're here to break down what just happened with Walmart and what's coming up with George Gilbert, former colleague, Wikibon analyst, Gartner analyst and now independent investor and expert, George, great to see you. I know you're following this space like you read about. Remember the first days when Databricks came out, we were talking about them coming out of Berkeley? Snowflake, we're all over Snowflake in the early days. We collectively have been chronicling the data movement since 2010. You were part of our team. Now you've got your nose to the grindstone. You're seeing the next wave. What's this all about? Walmart building their own SuperCloud. We got Bob Muglia talking about how these next wave of apps are coming. What are the super apps? What's the SuperCloud to you? Well, this keys off Dave's really interesting questions to Walmart, which was like, how are they building their SuperCloud? Cause it makes a concrete example. But what was most interesting about his description of the Walmart WCMP, I forgot what it stood for. W, Walmart cloud native platform. Okay. He was describing where the logic could run in these stateless containers and maybe eventually serverless functions. But that's just it. And that's the paradigm of microservices where the logic is in this stateless thing where you can shoot it or it fails and you can spin up another one and you've lost nothing. That was their triplet model. Yeah. In fact, and that was what they were trying to move to where these things move fluidly between data centers. But there's a but, right? Which is they're all stateless apps in the cloud and all their stateful apps are on-prem. Or the stateful part of the apps are in VMs. Okay. And so if they really want to lift their SuperCloud layer off of this different provider's infrastructure, they're going to need a much more advanced software platform that manages data. And that goes to the next thing. The Muglia and Tristan and Andy that you and I did that's coming up next. So the big takeaway there, George, was I'll set it up and you can chime in. A new breed of data apps is emerging. Yeah. And this highly decentralized infrastructure. And Tristan Handy of DBT Labs has a sort of a solution to begin the journey today. Muglia is working on something that's way out there. Describe what you learned from that. Okay. So to talk about what the new data apps are and then the platform to run them, this is, I go back to using what will probably be seen as one of the first data app examples as Uber, where you're describing entities in the real world, riders, drivers, routes, city plan, these are all defined by data. And the data is described in a structure called a knowledge graph for lack of it. No one's come up with a better term, but that means the stuff that Jack built, which was all stateless and sits above cloud vendors infrastructure, it needs an entirely different type of software that's much, much harder to build. And the way Bob described it is you're gonna need an entirely new data management infrastructure to handle this. But where, you know, we had this really colorful interview where it was like Rockham Sockham, but they weren't really that much in opposition to each other because Tristan is gonna define this layer starting with like business intelligence metrics where you're defining like things like bookings, billings and revenue in business terms, not in SQL terms. And those- Well, business terms, if I can interrupt. He said the one thing we haven't figured out how to APIFI is KPIs that sit inside of a data warehouse. And that's essentially- That's what he's doing. Yes. Right, and so that you can now expose those APIs, those KPIs that sit inside of a data warehouse or a data lake, a data store, whatever, as through APIs. And the difference- So what does that do for you? Okay, so all of a sudden, instead of working at technical data terms where you're dealing with tables and columns and rows, you're dealing instead with business entities using the Uber example of drivers, riders, routes, you know, ETA prices. But to that, you can define, DBT will be able to define those progressively in richer terms. Today they're just doing things like bookings, billings and revenue. But Bob's point was the today, the data warehouse that actually runs that stuff, whereas DBT defines it, the data warehouse that runs it, you can't do it with relational technology. Relational technology, caching, architectures. SQL, you can't do it. SQL caching architectures in memory, you can't do it. You've got to rethink down to the way the data is laid out on the disk or flash, which by the way, Thomas Hazel, who's speaking later, he's the chief scientist and founder at Chaos Search, he says, I've actually done this. Basically leave it in an S3 bucket and I'm going to query it, you know, with no caching. All right, so what I hear you saying, then tell me if I got this right. There are some things that are inadequate in today's world that's not compatible with the super cloud wave. Specifically how you're using storage and data, stateful, and then the software that makes it run. Is that what you're saying? Yeah. There's one other thing I want to, you mentioned to me, it's like when you're using a CRM system, a human is inputting data. Nothing happens till the human does it. Right, nothing happens until that data entry occurs. What you're talking about is a world that self-forms, pulling data from the transaction system or the ERP system and then builds a plan without human intervention. Yeah, something in the real world happens where the user says, I want to ride and then the software goes out and says, okay, we got to match a driver to the rider, we got to calculate how long it takes to get there, how long to deliver them. That's not driven by a form other than the first person hitting a button and saying, I want to ride. All the other stuff happens autonomously driven by data and analytics. But the... But my question was different data. So I want to get specific because this is where the startups are going to come in. This is the disruption. Snowflake is a data warehouse that's in the cloud. They call it a data cloud. They refactored it. They did it differently. The success, we all know what it looks like. These areas where it's inadequate for the future are areas that'll probably be either disrupted or refactored. What is that? That's what Mugley's contention is, the DBT can start adding that layer where you define these business entities. They're like mini digital twins. You can define them, but the data warehouse isn't strong enough to actually manage and run them. And Mugley is behind a company that is rethinking the database really in a fundamental way that hasn't been done in 40 or 50 years. It's the first, and his contention is the first real rethink of database technology in a fundamental way since the rise of the relational database 50 years ago. And I think you admit it's a real Hail Mary. I mean, it's quite a long shot, right? Yes. But it's huge potential. They are pretty far along. Well, we've been talking on theCUBE for 12 years and about 10 years going to AWS re-invent, Dave, that no one database will rule the world. Amazon kind of showed that with them. What's different? Is it databases are changing or you can have multiple databases or? It's a good question. And the reason we've had multiple different types of databases, each one specialized for different type of workload. But actually, what Mugley is behind is a new engine that would essentially, you'll never get rid of the data warehouse or the equivalent engine in like a Databricks data lake house. But it's a new engine that manages the thing that describes all the data and holds it together. And that's the new application platform. George, we have one minute left. I want to get real quick thought. You're an investor and we know your history and the folks watching George got a deep pedigree and investment data and we've been contested by against that. If you're going to invest in a company right now, if you're a customer and I got to make a bet, what does success look like for me? What do I want walking through the door and what do I want to send out? What companies do I want to look at? What's the kind of vendor do I want to evaluate? Which ones do I want to send home? Well, the first thing a customer really has to do when they're thinking about next gen applications, all the people have told you guys, we got to get our data in order. Getting that data in order means building an integrated view of all your data landscape, which is data coming out of all your applications. It starts with the data model. So today you basically extract data from all your operational systems, put it in this one giant central place like a warehouse or a lake house. But eventually you want this, whether you call it a fabric or a mesh, it's all the data that describes how everything hangs together is in one big knowledge graph and the different ways to implement that. And that's the most critical thing because that describes your Uber landscape, your Uber platform. That's going to power the digital transformation which will power the business transformation, which powers the business model, which allows the builders to build, coders to code, that's super cloud application. George, great stuff. Next interview you're going to see right here is Bob Mugley and Tristan Handy. They're going to unpack this new wave. Great segment, really worth unpacking and reading between the lines with George and Dave Vellante and those two great guests. And then we'll come back here for the studio for more of the live coverage of super cloud too. Thanks for watching.