 Hello, and welcome to the podcast, The Cube Pod. Episode 16, I'm John Furrier, Dave Vellante. Every week we break down what we've been working on. We've been traveling like crazy. Couple days at home this month, it's been a big road month for The Cube. Dave, great to see you. Episode 16, we said we go to 20 to see how it goes. So far it's going great. Got a lot of great feedback. If you like the podcast, drop us a DM. I'm wearing a shirt from a hot startup. Check it out. You want to send me a T-shirt? I'm happy to wear it. And we'll get the low out there. So Jenner of AI is the hype. Dave continue to go on. It's everywhere. The hype cycles on an all-time high. Guardrails are coming off. Oracle crushed their earnings. The EU lawmakers approved the Broadcom deal with VMware. Databricks is having a big event. Before they go public, they announced they're over a billion dollars. We think it's over two. And of course, SuperCloud 3 is coming up super fast. Our event, we have specials there. And Amazon had their big security event. We got some commentary there. Rob Bearden, friend of the cubes, leaving Cloudera. Yeah, I'm not sure what he's going to do. Reddit goes dark. That's a cultural moment. It will be in my rant section. A lot going on, Dave. Amazon's cloud unit considering AMD's new AI chips. NVIDIA buying a financial services company, financial software company for 10 billion exchange. Dave, a lot going on. It's a cultural shift. AI continues to dominate every week. Elon's kind of falling off the news cycle with Twitter. He's got a new CEO. She's digging in. But generative AI continues to hit every week for us. Every single company is just AI this. And people are afraid of AI washing. So you have people doing AI, not afraid to pump their own story up because they're afraid they're going to get labeled with using AI and then they actually have it. If you don't have some kind of AI story at your conference, people are like asking, why are you burying the lead? It's the most important story of the year. So you have to have AI in your narrative or you look like you're out of touch. I mean, I don't see how you couldn't stand up. Now, to your point, people risk AI washing and people saying, well, we saw this at every conference we've been to. Cisco didn't AI wash in my opinion. They basically said, hey, we're going to infuse AI in everything we do. They didn't like try to come up with some radical thing. The way Dell handled it, they said, hey, we announced Project Helix, which is this partnership with NVIDIA. And it'll turn into a product next year. And so, but every company has to have even AWS, although last week at Reinforce, they talked about CJ Moses, the CISO of AWS, brought AI, generative AI up in his first sentence. Although I don't think they really connected the dots at the keynote, but they definitely did in the analyst session. So I don't know about the journalist stuff that you were in, but they talked a lot about generative AI and gave a lot of really good explanations and examples. So they, I think tried not to AI wash in the keynote, but I do think they kind of missed the opportunity to talk about more about what they were doing in AI. I mean, it's interesting, Amazon Web Services, we saw my land just on Bloomberg, they're trying to say, well, you guys were late to the game and she's pushing back saying, look, no, we weren't late to the game. So, you know, because they're trying to, the press are trying to put Microsoft up against AWS on the AI front. When you got, Google's got shops too. So you got, you know, Vertex, you got open AI, you got, AWS has their version. So she was very clear. She's like, we've been doing AI for a long time. And I think people don't want to, you know, tooth their own horn, so to speak, as they say. And I think that's a mistake. And I think if you got it, flaunt it, but not everyone has it. So there's a lot of AI washing, accusation, meaning people throwing AI and everything to kind of seem hype. And we're seeing people doing it in our industry all over the place. I think being AI enabled, it's going to be the different, like to me, the characteristics we're going to see in announcements. If they're AI enabled, then they're going to be good. Because at the end of the day, there's two questions that everyone's asking. How do you build it? And how do I stand up the infrastructure to support it? And that fundamentally is a developer concept. That's what open source development's all about. And it's getting faster, smaller, more efficient and highly scalable with the cloud. So just this week, we had an announcement this look and angle and we had an exclusive with OctoML, a Madrona Ventures company, they're shipping and debuting self-optimized compute services for generative AI applications. We have any scales trying to do something different. So you're starting to see some of these data infrastructure next-gen clouds pivoting quickly to be a infrastructure service to stand up the builders who are trying to figure out how to build AI apps. And to me, that's the big story right now is that nobody actually has that playbook and it's gonna be, I think, a rush to figure out who's gonna land where with their developers. How do you build AI apps? Who do you use? How do you unlock that potential for generative AI? And it's gonna come down to how do you build it and how do you run it, right? So that's gonna be the first inning, if you will, in this game, in my opinion. So you start to see a lot of people in SaaS, Salesforce, so we got AI. Now Salesforce is totally AI washing, in my opinion. ServiceNow, I think they're AI washing, in my opinion. So, but they have a lot of AI enablement. So that's the kind of the distinction. Do they really have AI or are they enabled for it? So I mean, that's kind of a rant section right there, but still. Well, but I don't know if I totally agree with it. I mean, I think that I look at Salesforce and ServiceNow as enterprise applications. I would say this is similar with Oracle, although maybe a little different, but SAP, where they're infusing AI into their apps. Why wouldn't they? And they've been talking about doing this for years and now they're maybe hyping AI a little bit. Why wouldn't you though, like we talked about earlier, I don't look now though that all the AI fraught has pushed the stock market into bull market territory. And people, all investors are now like, is this a real bull market? The yield curve is still inverted, but you can have a bull market during when the yield curve has been inverted. It's happened several times. So it's really confusing right now, but it's like I said, I don't look now, but it's a bubble. I mean, it's a legit bubble. Okay, we just did a story on this month old Paris-based startup, Mestrel, I think it's called, Mestrel AI, raised 113 million in the seed funding, Dave. The seed funding. Like have you ever seen- That's official, we are in a bubble. That's bubble mania right there. That's very bubble-licious. So, you know, that's a bubble-licious mentality. But again, this is again the rush for the dot-com bubble kind of comparison we've been having on theCUBE here. But to me, there's legit operational data out there that's leveraging for AI. And this is where I think you're breaking analysis on Snowflake, we've got another breaking analysis coming up on Databricks. Those two shows going to be happening this month. One's a public company, Snowflake. Databricks is private, but they claim to be doing over a billion dollars per Bloomberg. We think it's over two plus billion closer to that number. But this brings up the question, Dave. What's the role of data, right? Because you have old school data warehouse and then you have this modern data kind of layer stack super cloud capabilities where operational data is much more aligned with the AI models because operational data is real time, it's more targeted towards an application. And the benefit of operational data is addressability and abstracting it away, managing it so that no one has to deal with it. And today operational data is viewed as boring, it's data, it's in a database. So I think you're going to see data tiers segmenting out around what's data for value, system of record and system of engagements, but this new operational data that developers are going to do. And I think, I look at MongoDB for instance, they just had massive earnings. They have what is a data platform, they have a database, but the developer model is the same whether you're an enterprise or an enterprise doing a hackathon. But you got production with Atlas and they have a lot of operational data that's tied to applications, not the big oracle, like big database. So I think it's very nuanced in the industry, but I think the data wars are going to become and quickly, it's going to be the approach, the culture and then ultimately who's going to take advantage of the best data architecture for AI. And that's where the Wall Street bubble will probably pop or grow more in my opinion. Well, you can't do AI without data. So it makes sense that these data platforms like a Snowflake or a Mongo, we know Databricks is not public, but if they were, they'd be getting a tailwind from this. Look at what happened with Snowflake. I think on May 24th, they announced earnings. They were conservative, they basically, we kind of were very conservative on the guidance, the stock got hit. And then I think people realized, oh wow, this is a buying opportunity. And then they jumped in and the stock rocketed through, it's a little off today, but I think once it hit 170, it was off and running. And so, and then Mongo, different, but similar. I mean, people, first of all, Mongo absolutely blew away its earnings. And what it did, if you look at the Mongo numbers, it essentially pulled forward its guidance by a year. So they accelerated by a year. I mean, that's like, the people, investors were like, holy crap, we got to get in. And we've been following Mongo since they were called TenGen. You remember TenGen? Exactly. At the time, there was criticism, well, they might not scale, they don't have asset compliance. And then Mongo did an amazing job of actually addressing all those enterprise needs. And then, and then combine that with the developer affinity and the simplicity of Mongo in a document, data store. And it's just exploded because people are building apps on top of Mongo. We asked David to cheer you here at last AWS re-invent. Are you an ISV? He's like, no, I'm not an ISV. We're more than an ISV. We are a platform on which people are building applications. ISVs are building on top of Mongo. They're building on top of Snowflake. They're building on top of Databricks. And that is the future of data platforms. Those are the guys who are going to really win big in this market. And you mentioned Mongo. They have 2% of the market share in databases, the classic TAM that the Wall Street analysts kind of call the database market. And you got the three categories. You got the legacy, oracles, you know, and the vintage kind of guys past 15 years. Then you got the more robust data infrastructure for like the snowflakes of the world. And then you got the old school oracles, right? So the interesting thing about Mongo, Dave, is that their first line of code was the year that Amazon really started going on. So then you called, that was called 10Gen. So around 2013, the growth of AWS and Amazon and MongoDB during those years from early days of Amazon up to about 2013, I think that's when they started working on Atlas to come five years to get Atlas. They actually had a scaling problem, but they were riding the wave of cloud just like Amazon had scaling problems or not enough higher level services. Because remember, everybody still uses EC2 and S3, whether you're a startup or an enterprise. MongoDB is the same exact kind of mirror image of Amazon in databases. Developers can still use MongoDB today and just build it for their app whether it's a startup and or an enterprise. So they've grown on the back of the cloud growth. And I think that's made them successful. And that's why I think MongoDB falls into that classic company that's outside the TAM of a database. So they can certainly nibble away at that 2% and make it or increase that 2% to a number. But I don't think that's the real market they're targeting. I think they're a developer first platform, Dave, because, and we're going to find out more in it next week when we're in New York, but they're a developer platform and the rise of open source, especially with generative AI, you need a database for any application, whether you're using a framework that's open source or not, you're going to need a database. So again, what why people still need EC2 compute is the same reason why Mongo is not going to go away. So if I'm an investor, I'm investing in MongoDB because they have headroom. I mean, the 2% number on the market shares is tiny. So there's room to grow there, but they're creating a category. They're like developer, they're developer first data platform. Like who else has that? So the thing about Mongo, because they've done such a good job of putting in all those enterprise capabilities that we were talking about earlier, they are operational, right? They're talking about operational workloads. And that's the other thing that's not really well understood about Mongo. It's not just about, for Mongo it's not just about the next customer, it's about the next workload. I think that's the mindset of Mongo just doesn't observe a watching Mongo over the years is they're going after workloads. They want to gobble up those new workloads and they're trying to expand the number of workloads that they can accommodate. And so that's their approach. Snowflake's different, right? Snowflake is analytic, right? And so started with data warehouse. You know, Oracle of course does both. They got one of everything. And then you got Databricks, which is traditionally that data science, machine learning, AI workloads. They're of course getting into sort of the data lake, data platform space. Snowflake going into the data science space. So they're on a collision course. And Mongo, interestingly enough, so you've seen like the consumption of cloud, it's sort of hurt Snowflake's earnings because they sort of follow. Snowflake is very much like AWS in my opinion. People dial down AWS, they dial down, they try to optimize, they're using Graviton, they're saving money. So it's different from Mongo. I think Mongo is less susceptible to that sort of consumption trend because they're running operational workloads. And that's the, I think is not well understood sometimes about Mongo. Well, I think that operational data I was saying earlier is valuable, right? That's more valuable to the AI than some of the other data. And certainly they're going to need to work together. And MongoDB is also has a cultural shift going on that's aligning with the marketplace. They want to abstract away the database from the developer, make it easy. Other people want to amplify up the role of the database, make it more, more usable when in reality developers want that under the cover. So in a way they want to abstract away. That's what software does. They're software database. So that's one. The other thing that I think about Mongo is if you know, the old expression we always say in the cube that Jeremy Burton would say, you never fight fashion. And I think Mongo's like a designer, designer brand. They're like, they're fashionable in every decade that they've been existing in. And I think now more than ever with open source they can appeal to the build it, use Mongo for anything kind of model, where you can use it on the hackathon in the enterprise or a developer. So they're very much like a designer brand. Like they're good, they're durable product. I think their earnings points to the fact that they're durable and continuing to grow. And by the way, like the cloud, when you get in, you grow with it. So I think that's the classic land adopt expand growth model. And it's very misunderstood. I think Wall Street, I listened to their transcript the other day, Dave. And from the earnings report, because they had amazing earnings, I think you quoted the line to the moon, Mongo to the moon. That earnings took a lot of people by surprise. And the Wall Street analysts were like scratching their heads, where does this come from? You only have 2% market. I don't think they even get it, Dave. So. I think the other thing about Mongo is, I mean, I'd say the same thing for Snowflake, but Snowflake was growing like crazy. Mongo never really went for growth at all costs, right? They said, hey, we're going to go steady as she goes. And I think, you know, Wall Street just, okay, they'll take you at your word. And so they weren't one of these growth at all cost companies, but now people are saying, oh my God, they've got to your point. They only 2% of this 80 to $100 billion marketplace. And they've invested in the future. They're developer friendly. They're really easy to use. They're picking off more workloads. So again, it's not just accounts. It's workloads in those accounts. Holy crap, we got to get in. You know, I got a nice compliment the other day from someone in the industry. They said, we love your pod. It brings back the old John and Dave days from when we started. And she said, you guys are always picking the right waves, the next big trend, before anybody else. And I think that's kind of continues to be the case. And with that in mind, I wanted to share an article with you Dave, get your thoughts, because in the Wall Street Journal, I just had an article, I think this speaks to, you know, separating the signal from the noise with AI and this gender of AI startups. A lot of hype. I mean, there's more hackathons now, more meetups than I've seen since the web 2.0 days. It's really hyped up. But if you want to see where the money is and understand who the winner and losers are going to be, here's an article I think that gives a little tell sign and maybe put that into the cube is right category in the next 10 years. The money won't be in the software, the gender of AI software. It's in the data. So the article is in the Wall Street Journal. AI startups have a ton of cash, but not enough data. That's a problem. So in the article, they're essentially calling out the trend that it's the data there. So as many of the old expression, it's the network, it's the dumb network stupid. It's the data stupid kind of thing going on. The people who have the data have the money. So if you ask yourself, who will make more money with gender of AI? It's the people with data. So the startups that are coming out that are getting massive valuations, they're building software mechanisms, which by the way, the barriers to entry are low, as you can tell, tons of startups. So there's going to be a fast rise and a massive decay of failed startups coming in if they don't tap into people with the data or that get the data themselves. Like, we're data full from a video standpoint. So we're probably in a better position to do say a video AI startup ourselves or someone who has an enterprise or MongoDB or big database in the cloud or data in applications will have an example. So what's your reaction to that? Because to me, I think this is a tell sign to where the successes will come out of the woodwork that no one will see coming. You'll see the glam, oh yeah, I'm doing gender of AI. Look at this great demo with some data, but the data scale, operational data or historical data, real-time data, whatever it is. It has to be the payload, the input. So I just emailed to Brendan, who's our show producer, the interview I did with Chris Lynch a month ago. It's all about the data, stupid, classic Lynch, right? And he didn't take off his shirt. I was happy about that, but his point was exactly the point you just made, that if you can have AI, but unless you have data, your AI is useless. And so, and I think your other point about real-time is interesting, today's breaking analysis, George and I interviewed Uber and how they built a real-time fulfillment system, which is incredible. I mean, you think about the Uber app and they had to rewrite the app in like 2014, 2015 in order to accommodate not only availability and low latency, but also consistency. So they didn't have necessarily consistency before they were using Cassandra. And the way Cassandra works is it's the last write-in that wins, even if it's not the most consistent one. So they had to rewrite, they had to change the engine while the plane was flying. And the point is it's all about real-time. Everybody wants, whether it's Kafka or something similar to Kafka, real-time streaming, that's to me the future, because apps or data apps are going to be all about taking your business and creating digital twins of the business, people, places and things. So it's not going to be about, serving up compute and storage. It's going to be about giving the access to the case of Uber, drivers, riders, destinations, ETAs, goods that are going to be delivered. And that's a whole new opportunity. And to your point about the data, you got to have data that data has to be coherent and you can apply AI to anything, but if the data's not good and it's not coherent, then what's the point? Yeah, and I think the success, it's going to be very easy for a startup or a couple of kids in a garage to do a demo with data and saying, look at this great AI, but the scale will come back into playing. This came up with the developer focus conversations we've been having. If you look at open source right now, if the developers can quickly build in, you've seen more AI apps on the long tail of open models coming in and open source, it's phenomenal. So we know it's fast to code. So that's going to be open for everyone. That democratization trend, I believe will happen. So you'll see a democratization of coding, but the question is, who's got the data? So that means the software app side of the business is the transformation focus area. So if I'm an investor or I'm a business leader, I'm going to be like, I need the app to be focused on the app, not the underlying mechanisms. It's got to be the application. And that's where I like the people that are thinking about making the developer productivity the same as we saw in security. You go back, remember seven years ago when we were doing theCUBE, no one talked about shifting left ever. Now it's shifting left to build security guardrails in there and good policy developers do it in their pipeline where they're coding, point of code. Data is coming fast to point of code. And I think this is going to be the operational data model, the time series, the real time, the valuable data, clean data. And I think it's going to be a whole data engineering practice dedicated to feeding clean data into the app for AI consumption. So to me, that's going to be a whole nother data infrastructure opportunity. Again, we're speculating, but the dots are easy to connect, right? Today operational data is boring, slow, whatever. But when you think about what it takes to control AI, it's the inputs, right? So garbage in, garbage out. No one, they don't want it to have to have that. So, no one wants to say that because they don't want to say that their data is garbage. Let's face it, most people have garbage data and they got to think about cleaning it up. So this is going to be a whole nother conversation but I think it's worth tabling now that the role of data and into the development process is going to be unpacked big time. Well, and I think it's worthwhile to think about the progression of how enterprise applications have been built and it's changing dramatically. You go back to 1.0, which was sort of back office and departmental stuff. And then you had the ERP trend, which SAP won, but it was companies like Ford that were able to say, okay, I'm going to take my supply chain, my financials, that was sort of all my enterprise data. I'm going to put it together. And of course it required a data silo to support that. And then a lot of people don't realize this but in the sort of dot-com era, what Amazon did is they sort of created the first real integration of basically built a custom platform to manage all that internal data. And what Uber has done now is they've externalized that to an ecosystem, that riders, drivers, et cetera. And so here's the thing, John, not everybody has like these engineers, the engineer we interviewed from Google today, I mean, Uber today was amazing, but not everybody has, you know, hundreds of engineers with deep expertise that can do this stuff and build this stuff. So I think with the next three to five years, you're going to start to see Uber-like applications emerge that people can buy off the shelf. And the big change is that these apps are going to be data apps. In other words, it's not going to be the data that's buried inside the app where the business logic lives. It's the reverse, the business logic is going to be embedded inside the data and people are going to build applications on top of that. And that's a big change, great time. Dave, that's great, great observation. I'll also add that not only data apps but data libraries. So you're thinking about software as library, calling a library to do something for code, you're going to see data kind of have that library-like reference. It's whatever you want to call it, prompt, just tune that. But that kind of thinking means that data is closer to the development point of coding for the developer than ever before. Explain what that means. Explain what that means, the libraries. What's the implication for developers? Explain that. Why does it simplify things? Well, if you have a operational data abstract that we say whether it's MongoDB or a database like a Redis Labs or something else, you want to call the data, right? But you don't want to have to get in the weeds of making a ticket call to go with some data science. It's just like your point about democratization. Data science went through the same thing. You got to be a really, really strong math person to be data science. And now it's like just running a spreadsheet. It's really easy. So they democratize data science. You're going to have that same kind of level here with this AI. You don't have to be a machine learning Uber engineer to code data. So I think what you're going to see is that the work, like a basic application developer, someone in open source who knows how to code is going to get assistance from the AI bot, human plus AI, to be more productive. So an average or strong developer, not a one percenter and machine learning guru. You're going to be able to code machine learning. You're going to see platforms develop. So I was talking earlier about the designer brand of Mongo. I think like clothes and fashion, you're going to see like you can have great designers. They all have different brand names, but it's still clothes. So you're seeing it now in security, Palo Alto Networks, they're building a platform. That's going to be a designer brand in cyber. They're going to have platform, they're going to have tool integration, maybe not best of breeze, but for the person, the company wants to wear that platform. They want to scale and grow. So you just have to see these designer kind of vibes going on where it's like, this company does everything. Mongo will do A to Z from developer to scale. You're going to see these platforms emerge. And I think that's going to be a new market dynamic. So, and that's going to empower developers to be democratized. So whether it's AI code or developer and open source, they can now write apps on top of these kind of platforms. That's where the democratization truly happens. And I think that's what data libraries, data apps will do. They'll be available to be coded. Meaning, I'm writing an app, I need data. It's got to have domain specific value. I need horizontal scalability. That's cloud. So cloud is enabling this new platform, replatforming of companies. And it reminds me of a certain kind of fashion, the designer brands you got. Versace, you know. You go. You go. You go. Everyone will like their certain designer brands. And I think platformers will be brands. And I think that's going to be interesting to see. And even well established companies are already scrambling to prove that they're going in this direction. Look at Centure plans to invest three billion into AI. They're a brand. They're a designer brand in a way. So it's very interesting. And once you pick, you're going to have the same. So I think it's not about proprietary. It's more of, can you do it? Can you get the job done? And in developer world, it's, can I scale? Do I have the quality of service? Do I have all the things I need to support that growth to run AI? So it's going to be very interesting to see. And I think the data equation is not yet figured out. So today the problem is no one knows what to build on and no one knows what to stand up for infrastructure. So this is inning one. Inning one will solve those two problems. And that's what, and everything else is hype, right? So everything else is hype, except for the startups that come out to, to those things. In content, you see the large language models for video, texts, that's kind of more kind of a different personal thing, but same thing. And again, you got to have the data. You got to have the data. So changing subjects a little bit. You saw AMD announced its GPU, its AI chip earlier this week, but it's not going to be available for a while. And you know, this is important because the world needs an alternative to NVIDIA. The problem is NVIDIA is so far ahead. You know, it's Andy Jassy's, there's no compression algorithm or experience. Exactly. And NVIDIA's CUDA architecture has, you know, they made a huge bet on that, you know, years ago. I don't think Wall Street liked it at the time, but they're loving it now. So you got AMD, you have Intel, I was, I shared with you, I think on an earlier pod that I had dinner with an AI guru out in Las Vegas at one of the events. I won't even say who it was because I don't want to give this guy away because, and he was basically saying, we love Jensen. He said, because he's a baller, but we need alternatives because basically NVIDIA is gouging us and he said, you know, AMD's coming Intel, they actually use Intel GPU chips today, AI chips today. He said, but they're so deficient, but they have to use them because they need alternatives to NVIDIA. So NVIDIA's lead is substantial. And a lot of people are saying, well, you know, that lead could get compressed, but I don't know, John, I think NVIDIA is going to maintain that lead for sometimes they could, they could potentially become the AI monopoly. Even though they didn't get ARM, and that's the other thing, Intel's supposedly investing in ARM, I guess Intel's becoming a, I don't know, an investor now. Intel missed the two biggest ways of our generation in terms of compute opportunity. So certainly around GPUs, they missed the whole crypto mining market, which was massive for NVIDIA. And now a large language models and AI. So, you know, I think it's still time. And they miss smartphones, right? I mean, that's, it goes back, but I sometimes. Well, they're legacy, right? They're the legacy brand. But sometimes I think people think I'm like rooting against Intel, they're far from it. I mean, the United States needs Intel to be successful. It's just that they have such huge challenges and, you know, trying to do foundry and design and compete is really, really hard. So it's interesting that they're supposedly going to be investing in ARM's IPO. It's like, okay, you know, great. But that's not going to save Intel, right? It's a nice, smart, I guess. And I like the fact that they're actually going to be in their foundry building the ARM designs. Intel stock price is directly correlated to the fact of how many cube appearances Pat Gelsinger has. And right now, right now it's zero appearances. So, Pat Gelsinger, get on the queue there. I'd love to have Pat back on. He's a friend and I mean, I think he's the one person who might be able to save this company. You know, kidding aside, which by the way, Pat needs to come back on the cube, is that Intel still has massive share. I mean, look at what they're doing. I mean, half the market, that three quarters of the market doesn't even need the high-end stuff. I think it's easy to throw rocks at Intel and say, you missed that, you missed that. But that's just kind of the bad, the aircraft carrier that is Intel. It's hard to be that nimble. Nvidia made good calls early on this stuff. They were early on the GPUs and they made an investment, billions of dollars. So it's not like they were Johnny come lately and decided to be dominant in the GPU market. They made a bet and it paid off. And I think the other bets weren't made properly and there it is. So, you know, I think Intel, and everybody has a chance in the GPU. Again, you're going to quote, you know, Amazon, Andy Jassy, I'll quote Jeff Bezos. Your margin is my opportunity. Is that play here? And I think in the GPU market or whatever the compute market emerges into with AI and other use cases this time. So, you know, I think there's going to be an opportunity to get back in the game, especially if Nvidia holds the line on their pricing and bogarts the market. So, you know, notwithstanding geopolitics in the manufacturing areas around the world, there's a chance that everyone's going to be back in this business and there's a lot of margins. So, Nvidia could play this properly and manage their pricing and their supply chain to keep that competitors off balance, keep those competitors off balance, but we'll see. We'll see what happens. I mean, the challenge for Intel as we've talked about is learning curve and you got foundry. So, Intel does both with those you don't know. Intel designs chips and they manufacture chips. They have an integrated approach and integrated manufacturing design and manufacturing approach. And it served them so well for so many years. But the problem now is when the industry sort of said, hey, we're going to sell off all the foundries, right? AMD, for example, sells its foundries. IBM got rid of its foundries because even though, you know, IBM still designs semiconductors. They design the semiconductors and they ship them out to foundries like TSM, like Samsung, now like Intel to manufacture the chips. The problem is learning curve, experience, volume. TSM by far number one foundry on the planet, thanks largely from their deal with Apple. They do massive volume with Apple. Samsung kind of number two, but also very, very good. State of the art manufacturing and Intel is sort of a distant third right now. And so the problem is it's called Wright's Law, which basically says that your costs basically get cut when you double and only when you double your volume. So if it takes forever to double your volume because you're tied to PCs and PCs aren't growing like smartphones, then you can't take advantage of the economies of scale. And that's where Intel is really struggled because it passed on the smartphone. And so how do you recover from that? Well, you got to do good design. You got to spin out the foundry. You got to try to get some volume. And I think we're seeing some signs that it's this hope. But as far as the delta between TSM and Intel's foundry, it's just a massive difference. And I was also say this, it's interesting that Warren Buffett invested in TSM and then he pulled his investment. And the reason he gave is because of the geopolitical concerns over China potentially invading Taiwan. Well, those same concerns should apply for Nvidia and anybody who's using an Apple, anybody who's using the TSM foundry, that could be some serious disruption. So that's why we have to root for Intel. We want them to make it here. We want them to succeed. It's just the hurdles are enormous. If anyone's going to do it, it's Pat Gelsinger. Speaking of kind of old garden and older than Gelsinger is Larry Ellison. He's still one of the giants remaining in the industry. He's just, you know, powering away won't give up and it's paying off Oracle. And if you look at their stock over the past, say, seven years, Dave, it's up. You look at that just beat expectations and they reveal the generated high service and all, yeah, all surprises everybody. Okay. Oracle cloud is incorporating the new AMD EPC chip too. So processors in the cloud. So, you know, don't count out Oracle. And so, you know, this brings that, that the dark horse up to the close to the pack. I mean, I had Oracle way back in Google just in third from eight Azure, but you got to put Oracle in the line with the horses on the track, Dave, and Google Oracle right there, different companies, but Larry's powering away and he's been slowly padding the CapEx, more CapEx, more data centers. So, you know, Larry, he said it originally years ago. It's just servers and storage in the cloud. I hope Instac Ray is listening. Did you know he was breaking my balls at the... I saw him last night, actually. So, I hope you had this conversation. So in the first super cloud, it was the first one? Yeah, it was the first super cloud. I was debating that Oracle was an example of sort of emerging super cloud with the deal that they did for Microsoft. And his comment to me was, yeah, you lost me at Oracle. I'm like, oh, the classic VC wants to sell all his companies to Oracle, but now he's shitting on Oracle. But to your point, Oracle's now at over $330 billion market cap and we talk about a misunderstood company. When Larry Ellison, I got an argument with, I won't mention his name, but it was a senior executive from IBM way back in early part of last decade when Oracle bought Sun. Oracle paid like $7.5 billion a Sun. And the net of cash was like, they got them for like $5 billion, $5.5 billion. And Safra worked her magic. And she took a company like Sun, it was getting one to one and a half, maybe two times revenue. And Safra, through her financial magic and the integration work that she did, they took that in record time, maybe it was over a course of 18 months, let's call it, took that $5 billion and instantly turned it into probably $40 billion a market cap. And that's set for Oracle on the engineered system route. Remember, Oracle and HP and UNIX, that used to be the standard. Well, Oracle said, you know what, Exadata, we don't need you anymore, HP, we'll do it on our own. And they set out to really invest in integrated hardware and software, basically the iPhone and the enterprise. And we've covered this for years, you interviewed Mark Herd, God rest his soul, and a number of other executives, Juan Lloiza, talked about engineered systems. And nobody beats Oracle at mission critical apps and nobody beats Oracle on Oracle. And the other thing is that people don't understand is Oracle invests money in R&D. They didn't get the cloud right at first. They didn't really poo poo it, but they did say, hey, it's operating systems and databases and memories and storage. And we know all that stuff. And now that's starting to prove true. I'll give you another little tidbit, talking to Uber again today. Who's Uber using? They have their own private data center. They run a managed service using Google Spanner. They use a lot of different databases. And they're using OCI, Oracle cloud infrastructure. So really interesting to watch Oracle hit all time highs, continue to invest in R&D. And you've got a founder led company who, like you said, never gives up, are at a war, and it's pretty awesome to watch. Oh, are you done? I fell asleep there. Are you kidding me? I'm only... I want somebody to time Dave's time on the- No, no, the Oracle, the Oracle, Oracle Congress, You lost me at Oracle. You lost me at Oracle. No, I was just kidding. I won't say it's a snoozer, but I mean, look at Oracle's boring, in a boring way, adding brick by brick to their cloud. And then their strategy is very larry-esque. Screw everybody, screw the fashion. We're going to just go do blocking and tackling, put our cloud up there, and we're going to go head to head on performance and latency and capability. The challenge I have for Oracle is the apps need databases. And you mentioned Spanner and a lot of other things. So yeah, it's going to be a code. This is an Oracle has to now figure out what they are to the modernization plan of an enterprise. And if you look at all the trends that's growing with cloud native, as is the application diversity, is you could run a database, whatever you want. You'd have a time series for this use case. Maybe you use an object store here or a document database here. I'll run Elasticsearch here. I'll run MongoSearch there. I got Redis over here, Cockroach Labs over here. The apps will have their own database choices under the covers. So the question is, what does Oracle, what's Oracle's play there? And if you look at it again, back to the database market share, Mongo's 2%. Okay, now Mongo's not a big company compared to say Oracle, their footprint's tiny. But is it really a database market, Dave? Or is it a platform market? So again, this comes up as Oracle position to win the modernization wave. Everyone's moving off virtual machines. Yeah, they want high-powered infrastructure. But is Oracle going to win those new modernization? Or are they going to get other databases? No, this is where Mongo beats Oracle. Because Mongo developers want to develop apps on Mongo, not Oracle. And except that, you know, we're talking about how Intel missed a bunch of trends. We could talk about IBM missed cloud, IBM missed AI, now they're trying to regain AI, IBM make the believe. Because they fumbled AI. They fumbled AI. They fumbled AI and somebody else is picking up the ball. And now they can get back in, but IBM could have dominated AI. What has Oracle missed? They didn't miss cloud. They're all over AI. They've been doing, you know, with Autonomous, they've been doing. And so my point is that Alisson is a tech, he spends money on R&D. And Oracle are very good at trans, unlike IBM, at translate, IBM's getting better, at translating their R&D into product. IBM was horrendous at that for years under Ginny Rametti and Palmasano. Arvin much, much better at focusing their R&D and turning it into product. But if you don't turn your R&D into product in the technology business, then what's the point? And that's your Xerox PARC, you know, or Bell Labs, right? Which is now owned by Nokia. And those are all wonderful things. But for a company to not monetize its R&D, that's a sin. Well, also, the Oracle has done a good job protecting their core business and they have masterful field and licensing deals that people complain about because they get locked in. But let's talk about another company. That was IBM's business too, by the way. IBM with DB2 was as functional and as mission critical as the Oracle database. IBM gave that up, lost that when they became essentially a services company. So you got to give chops to Alisson. You know, we always talk about Michael Dell, founder led companies, Oracle. Larry's Alisson is a prime example of that. You know, I think another dynamic, Dave, is that R&D's one of them, you pointed that out, well, I think some companies just lose the ability to compete effectively with competition. Because you have stakeholders, you have customers. Yeah, customer centric, we're like, you know, customer obsessed, great. But at the end of the day, it's also competitive. Oracle ate IBM's lunch in database. Okay, that's a fact, right? IBM's does a service there, okay, whatever, but they didn't win, Oracle won that one. IBM had to compete, they didn't compete, right? VMware just got the approval from EU for the Broadcom acquisition to go through. Our Broadcom got the approval for the VMware. VMware was trying to do R&D with Tanzu and Project Monterey and all the things they were doing with Cloud Native because as virtual machines go to containers and Kubernetes and microservices, VMware clearly saw that R&D, but Broadcom bought them. So will Broadcom keep that with VMware? So we're going to see that play out. This week, the EU reportedly set to green light, Broadcom's $61 billion VMware acquisition. If that goes through, we were speculating that might not go through, but looks like it's going to go through, which means it's going to happen. And the US will let it happen. So we're looking at the fall timeframe for probably VMware to be completely part of Broadcom. Well, I'm not sure Broadcom is going to keep that R&D. We'll see what happens. Well, I have an opinion on that, if you're interested. Yeah, I am interested. So first of all, I asked Pat Gelsinger when essentially Dell was in the middle of sucking all the cash out of VMware, restructuring, doing the tracking stock, going public, all the restructuring its balance sheet. I asked Pat Gelsinger, do you have the necessary resources in R&D to compete? And he said, yes, absolutely. I question that. I think VMware, had it been able to redirect those resources into organic R&D, it might be in a different position, assuming it could spend the money in the right places. I think what Broadcom is going to do is going to much more narrowly focus VMware on a roadmap that is sustainable. Not that VMware didn't have a sustainable business, but they were trying to maintain relevance by going off into all these other different areas. I don't think Broadcom gives a shit. I think Broadcom cares about, I'm going to run a sustainable business with legs and I'm going to redirect my R&D more narrowly to that roadmap. And so that's going to be the difference. They're not going to stretch it thin. They're going to really focus it. And then where's that focus going to be? It's going to be on extending the core platforms. It's going to be on making NSX and vSAN, make those integrated, extending ESX, and Tanzu's key for multi-cloud. And I think all the other peripheral stuff is going to be up for rationalization. And I think the determining factor is going to be how much roadmap and mining are we going to be able to do from the install base if we redirect the R&D. So I think that's Broadcom's sort of DNA. And I think they will. I got into a little bit of a debate with Keith Townsend on this who said he felt like VMware would be better off as an independent, at least better for the ecosystem. And my only question on that is, is would they be able to maintain relevance if they did that? So there's this, you know what I mean, John? These companies, they want to be relevant, just like EMC wanted to be relevant. So they would go and buy companies and look what happened. I actually think that Broadcom will bring a governance, which is going to be good, maybe not so much for all customers to want to be all things. But I think it's going to be, it's kind of like Oracle. Is Oracle, everybody shits on Oracle. But at the end of the day, you talk to Oracle's customers, they get business value. And it's not worth them leaving the platform. And that's Broadcom's challenges. I think they'll succeed is they got to make sure that people don't want to leave the platform and they make a good business case to stay. Well, this is going to be a key point. I agree with you, Sid. I'd add to that, the tell signal for Broadcom will be immediately apparent. CA, computer associates, CA, they bought them, wasn't as strong of a brand as VMware. VMware's got great brand, they got a great community, they got great customers, got a lot of customers. But the challenge that we're seeing is we're seeing it with SuperCloud. We're seeing a lot of the VMware ecosystem migrate to our narrative with SuperCloud. Why? Because it's cool and relevant and it makes sense, it's the future. I've heard that Hocktan's buying into this multi-cloud connective tissue layer that VMware is trying to connect into, because there is a migration from VMware, VMs to microservices. That's not going to be a hundred percent migration. They'll still be great use cases for virtual machines on bare metal and other use cases. But clearly the modern era is here. And if you're a VMware customer or a user or an operator, you want to be relevant. You don't want to be looked at as old and dusty and like, oh, they're the VMware guys. So I think they keep the marketing team and we'll see that, they'll keep the marketing team. They might jet us on the event, maybe outsource that. I think they'll keep vSphere and make VMware relevant and cool. And I think that's going to require Broadcom to invest in marketing. Now that's going to be interesting. If they don't do that. But Tom doesn't do marketing. Well, this could be a first because VMware has a marketable brand. So they're marketable, they have a community. If they don't do that, if they don't do that, it collapses into CEA and we know that playbook, that's rationalizes word for cuts. I guess they have a rationalized business. That's called killing the business unit. Well, I guess they haven't had to do marketing because they're selling chips, but maybe you're right, maybe in software they're going to have to do more marketing. Hey, we've been reporting that Silicon is the advantage. You're seeing the physical layer, AI is coming in from the physical to the top. We interviewed some Broadcom people about the 50th anniversary of Ethernet. Ethernet's open, it's versatile, it's got an ecosystem. So I think Broadcom, I've always said this, maybe I'm kind of out there, but I think Broadcom could have a great front brand from the front of the house brand with VMware that they never had before and vertically integrate some of the Silicon into the application where we are reporting also, AI has huge advantages. So talk about data. Broadcom and VMware combination would have tons of workload data, both legacy and modern. If they buy into that, then AI is perfectly hit for that. So if I'm advising Hoctan, I'd say to him, look, you got to evaluate the gettable nature of what AI could do for VMware and what Silicon advances could happen for VMware. And if that's the case, you've got a clear line of sight, and of course, VMware had a relationship with Dell, a big OEM of Broadcom. VMware has a relationship with HPE, a big OEM of Broadcom. VMware has relationships with all their customers anyway. So this is a data center meets modernization opportunity. And I was a little bit out there, but it might be a little bit too far of a bridge to cross, but I can see that path. This is a really interesting point you're bringing up. So I think multi-cloud, cross-cloud services, no brainer for VMware. They can do very well there. I'm not surprised Hoctan is buying into that. I think there is money to be made there. My big question is what happens to Project Monterey, which is essentially VMware's version of Nitro, which is, it's really AWS's platform for virtualization and intelligent network controllers and it's all security structure. So that is, we've written about this. That's one of AWS's secret weapons. What happens to Project Monterey? Because Broadcom obviously has silicon design shops, and that is sort of the future and a potential reason to not leave your on-prem applications and put them into the cloud. So that gives the whole entire ecosystem an advantage that would be really unique for VMware, but it does require R&D. And so, but again, Broadcom has those shops. That to me is one that's really worth watching. The companies like Dell and HPE, who are big Broadcom customers, big VMware partners and resellers, I think they need Project Monterey. It's going to be really interesting to see what happens there. Nobody talks about it much. It's sort of in the bowels, but that is something that is, in my view anyway, critical to the future of traditional legacy companies competing with cloud. Yeah, I think there's a scenario, you're right. And no one's talking about the low-hanging hot take with the hot take is, yeah, the rationalize. I think that's knee-jerk reaction and that's certainly a scenario for Broadcom, but I think we have to watch this upside potential. And I think I would push for the VMware brand, okay? I would definitely integrate super cloud into the narrative for sure. And I would essentially integrate quickly a vertical and silicon physical layer to app layer and for modernization and lock-in and all the OEMs together surround the castles, so to speak. Dave, I mean, other, you know, coming to our top of our hour here, but some things to talk about, like we've got a big data week, a Databricks event, data plus AI summit, same week as Snowflake. You're going to be at Snowflake. I'm going to be at Databricks. Everyone's kind of saying, well, Dave's a Snowflake guy and John's the Databricks guy. Well, I'm like, well, I mean, not really. I'm a big fan of Snowflake. Yeah, not really. I like Databricks too. I'm a big fan of Databricks too. I love open source, you know that. So if I'm going to lean towards anyone, I love open source. I think it's hard to bet against open source. The question is though, is Databricks going to be able to hide the ball as a private company of Snowflake's out in the open? So I think Snowflake's got more execution chops right now because they've been out in the public offering. Can Databricks execute? And you know the difference between public company, you talk about all the time on your breaking analysis. You know, private companies can operate differently than public. So when I go to their summit, I'm going to be there with theCUBE. We're going to do some in video interviews there with theCUBE. I'm going to be looking around. I'm evaluating, can Databricks execute? What's their story? What's their AI story? Clearly it's going to be a lot of large language models. We know Dolly's a big part of their deal. But what's their strategy? Like what's going to, where's the meat? Because if they can't execute, Snowflake has to manage. I want to just poke at this open source thing. Because everybody's like, well, you know, open source will win. I think it's a real examples where you can, you can accommodate open source and you can still, you know, win big time with your sort of proprietary system. And I think that's what Snowflake's trying to do. I mean, look at AWS, you know, look at Oracle. Oracle will tell you they're open and they do use open source and MySQL heat wave, but even though that's sort of different. But any rate, my point is this, if you look at what Snowflake's doing, they're not just doing things like iceberg table, Apache iceberg to check a box and say, okay, we're open to, we can do open source. No, they were, I think Snowflake recognizes that certain data types from the open source community are going to land in iceberg tables. And they need to be able to natively interact with that data that's in iceberg tables, update it as well. And I'll have to move data around. And their strategy is going to be the best place to build applications. That's what Snowflake wants to be. And you can't do that without an open source. So they're going to do, they purchase companies that do Python libraries. They're going to have tons of open source just like Amazon, as I've said many times, Snowflake's a lot like AWS. It's like Amazon has a ton of open source. You can invoke open source all over the place in Amazon's console. I think the same thing's going to occur for Snowflake. The big question is, Databricks has a ton of momentum right now. And Databricks trying to get into the Snowflake's wheelhouse, Snowflake's trying to get into the Databricks wheelhouse. Which one's going to have an easier time of doing that? Because right now, there's tons of overlap. I've written about this between Snowflake and Databricks accounts. And why is because people have realized, well, if I want MLAI, I got to go to Databricks. If I want simple analytics, I'm going to Snowflake. And both companies have said, okay, we see each other and each other's accounts. We're going after that. We're going to make it easier for our customers not to have to go elsewhere. And that's the big battle that's going on right now. Yeah, I think I love Snowflake. I love Databricks. I love MongoDB. Companies like that, they sit on AWS. That's how they made their bones in the business. All of them grew out of basic functionality, keep on adding more enterprise features. While still being the same use case, what got them there? I mean, if you look at AWS and all these successes, and I was just having a conversation with someone about Mongo this morning about this one point, their use case workflow is the same as it was when they started. For developers at Mongo, you want a database, you can get it. As you grow, you get more. That's more of a, again, back to the designer brand, Snowflake. It's the same use case you had when you first started using Snowflake, but yet you got more. Databricks, same thing with Spark. Now they got Dolly. It's just everything's additive, Dave. And I think this trajectory is the super cloud. This is, again, super cloud is not just multi-cloud layer. It's how the developers are responding. There's not a lot of headfakes going on. The developers can stay on a track and grow and not have to be specialized, more democratization hitting the thing. So I think open source is the way it's proven it. In fact, open source has changed so much. It's not like it was before. I mean, it still is the same. It's still groups of communities where you code, but that's not what it is anymore. Now it's the industry. So what's happening is open source is about the experience of building. If you look at what's really going on in open source, it's about not just communities of free software. It's about how to build and be better. And so there's a lot of collective intelligence. There's a lot of frictionless communications. And so that's why I call it like fashion because once someone finds out the best way to code a China AI app that's lightweight and high-performance and how to stand it up with infrastructure, like with OctoML, everyone just does it. It becomes the fashion. It's organic, right? So developers now are driving the standards. And this is what Instacurea actually said on our super cloud. The developer, and he's right, the developers are setting the pace and creating the fashion or workflow. So it's not about a vendor, right? Saying this is the best solution. It's the organic. It's the people with the data. It's the people that can do it, right? So to me, that's the new open source dynamic. It's very fashion-like. It's like, here's the new thing. And it works. And everyone does it. And by the way, that's not influenced by any vendor. So if the vendors don't get in and contribute, they can't even play in the game. Well, to me, it's influenced in this way, John. Tell me if you agree or disagree. The winner, the winning platform is gonna be the platform on which it's easiest and fastest and most productive to build data-driven apps. That you have to be the best, the most performant, the fastest, the best open source tooling available to build applications, to be the best place to build applications. Whoever does that, whether it's Snowflake or Databricks or AWS or Microsoft. And it's not a zero-sum game to invoke Matt Baker. But that's the game. You've gotta be the best place, the number one place to build apps. That should be the North Star for all these companies, in my view. Absolutely. I think you're gonna see a lot of people building their own apps, solving their own problems, selling it to other people. I think the business models will change. It's gonna be very interesting. And I think the cultural shift, you know, I've been saying this for a long time, Dave, and I think open source is one illustration of how the culture is changing. Another one is how people build apps. And one of the things that caught my attention, I guess I'll start the rant section, Reddit went dark this week and everyone was going on a blackout, all the top threads, subreddits. Now, and then the CEO just had an internal memo, oh, we're gonna get through this. And he was quoted today as saying, our API was not in design for people to build a profit on top of. And so they're changing their terms of service on the API, charging an anxious amount of money. Brendan, our producer, was pointing out. This to me speaks to the culture. Reddit's going back to fourth grade with this move. They should not bite the hand that fed them. Their users are, as their company, that's how they got here. And the API dynamic is how people code. So if you look at the lingua franca for building apps, APIs are a central part of it. So why wouldn't Reddit embrace their crowd, their tribe, that's the young developers, let them profit on the API. Reddit should participate in that. They should encourage it. Instead, they're blocking it. So they're like going back to fourth grade. Like this is like- Very Twitter-like. Very Twitter-like. This is very backwards. No, it's very VC-like or very much like clueless type. So, yeah, it's like, again, do you wanna fight? Well, Twitter was pretty clueless with its developers. I mean, we lived it. Every single developer in open source, every single developer that's under the age of, say 40, even 30, if that matter, is gonna be like, this is crazy. Why would someone not let us build on top of it? So I think it's gonna set a real stake in the ground for Reddit. It'll probably be just another website. And I think Twitter blew this years ago before Elon took it over. When Dick Costello was the CEO, he made a blunder of Epic Proportion because he went more, wouldn't it be like Google? It versus being more of a developer-friendly company. So it was Dick Costello that actually blew it, Twitter in my opinion. So, well, he was a CEO in charge at that time. So Reddit CEO really blew it to you, Dave. And I think- Not developer-obsessed, this deal- Well, they're just tone-deaf to what people want right now. And that's, again, back to the open source dynamic and culture shift. And again, you're seeing a commercial culture shift by the enterprise vendors coming in to participate. So I just find that a bad move, that's my rant for the week. So my rant, sort of a shorter rant, I guess, as you saw in the news earlier this week, that it just kind of hit Monday, I guess, that Andreessen, and maybe it was kind of last week's news, but I don't think we talked about it, Andreessen setting up a crypto presence in London because the US government is trying to kill crypto. They're in bed with the banks. They don't like crypto. The Gary Gensler says, we did talk about this. Why do we need another currency? We have the US dollar. We have the Japanese yen. We have the Euro. We have the British pound sterling. Why do we need another currency? And you know my thought on this is, well, we need another currency because all you guys do is keep printing more money and spending more money. And so, so Andreessen is seeing that the trend is not their friend in the US. So they're going to London to set up crypto shop there because they think London can be the epicenter of crypto, which would be awesome. I was kind of surprised a little bit at London. I had to research this some more because, you know, with Brexit, but it seems like Andreessen's, I'm sure done the homework and feels like London could be the place for this. Big mistake by the US government pushing out the innovation, pushing it outside the United States. I'm not in favor of that. And so, I've always come back to this public-private partnership. I think it stinks with tech, you know, the Microsoft Activision thing, US back-during certain deals with the UK competitions authority and getting them to do their dirty work for them. I just, we need a better public-private partnership. Big tech is not all bad. If they're breaking the law, you've got to moderate them. There was an article this week somewhere about how the EU wants to force Google to break, you know, spin out his ad business. Okay, maybe there's some merit to that, but it's just like big tech is under attack. And I feel like we got to defend digital a little bit. And every now and then, think about the good that tech has brought to the world. And that includes crypto. Yeah, and I think you and I, you know, we've been talking since the pandemic started. Cube plus digital has always been our focus. I think this really highlights everything's plus digital is a big part of how the physical and digital world's coming together. And again, software, I'm a big believer software is going to be a big driver of that. So software power dynamics will start influencing things like data infrastructure, government, look for software truly to start eating the world in a way that's never been seen before with the kind of cloud scale. So, you know, we're going to continue to monitor with the Cube. This is one of our lenses that we're looking through. Obviously, the initiative AI is top front and center foundation models, innovation, new applications, infrastructure, security, crypto is one of those and blockchain and whatnot. So, you know, developers are in charge, how to operate things, that's the focus. Look ahead next week, Dave HPE Discover, MongoDB in New York City on MongoDB World or they're not local to doing a big kind of tour kicking off in New York City will be there. Week after that, Snowflake Summit in Vegas and Databricks Data plus AI in San Francisco. We're going to be busy. And of course, we've got a lot of virtual digital events happening and programs were rolling out. So, look for new packaged products from the Cube. You're starting to see these kind of targeted content pieces we're doing, unpacking AI, unpacking infrastructure, unpacking cloud. You're starting to see a lot more targeted programming. So, we're going to continue to keep pumping out the content Dave, because content is data and data is what's going to be key for this next gen AI. So, let us know how you think about it. Don't forget SuperCloud, don't forget SuperCloud. July 18th, we still got nominations. We got some speaker nominations coming in from the VMware and Microsoft Ecosystems. Very interesting, Dave. So, people thought leaders want to present like we did at DockerCon. So, a lot of sessions, we might introduce sessions on July 18th. So, AI meets security meets cloud. So, the bumper sticker theme. All right. Let us know what you think. We're on DMS, we're on all channels. We're open Twitter, Facebook, LinkedIn, Signal, WhatsApp. We're all there. Text us, email us. For Dave Vellante, I'm John Furrier. Thanks for listening, episode 16.