 Welcome back to SuperCloud 2, an open industry collaboration between technologists, consultants, analysts, and of course practitioners to help shape the future of cloud. At this event, one of the key areas we're exploring is the intersection of cloud data and now building value on top of hyperscale clouds and across clouds is evolving, a concept of course we call SuperCloud. And we're pleased to welcome our friends from Enterprise Technology Research, Eric Bradley and Darren Brabham. Guys, thanks for joining us. Great to see you, we love to bring the data into these conversations. Thank you for having us, Dave, we appreciate it. Yeah, thanks. You bet, and so let me do the setup on what is SuperCloud. It's a concept that we floated before re-invent 2021, based on the idea that cloud infrastructure is becoming ubiquitous, incredibly powerful, but there's a lack of standards within the, you know, across the big three clouds that creates friction. So we defined over the period of time, you know, better part of a year, a set of essential elements, deployment models for so-called SuperCloud, which create this common experience for specific cloud services that, of course, again, span multiple clouds and even on premise dates. So Eric, with that as background, I wonder if you could add your general thoughts on the term SuperCloud, maybe play proxy for the CIO community because you do these round tables, you talk to these guys all the time, you gather a lot of amazing information from senior ITDMs that complement your survey. So what are your thoughts on the term and the concept? Yeah, sure, I'll even go back to last year when you and I did our predictions panel, right? And we threw it out there and to your point, you know, there's some haters, anytime you throw out a new term, is it marketing buzz, is it worth it, why are you even doing it? But, you know, from my own perspective and also speaking to the ITDMs that we interview on a regular basis, this is just a natural evolution. It's something that's inevitable in enterprise tech, right? The internet was not built for what it has become. It was never intended to be the underlying infrastructure of our daily lives and work. The cloud also was not built to be what it's become. But what we're at now is we have to figure out what the cloud is and what it needs to be to be scalable, resilient, secure, and have the governance wrapped around it. And to me, that's what SuperCloud is. It's a way to define operatively what the next generation, the continued iteration and evolution of the cloud and what it needs to be. And that's what the SuperCloud means to me. And what depends if you want to call it meta cloud, super cloud, it doesn't matter. The point is that we're trying to define the next layer, the next future of work, which is inevitable in enterprise tech. Now, from the ITDM perspective, I have two interesting callouts. One is from basically a senior developer IT architecture and DevSecOps who says he uses the term all the time. And the reason he uses the term is that because multi-cloud has a stigma attached to it when he's talking to his business users. The stigma is because it's complex and it's expensive. So he switched to SuperCloud to better explain to his business executives and his CFO and his CIO what he's trying to do. And we can get into more later about what it means to him. But the inverse of that of course is a good CISO friend of mine for a very large enterprise says the concern with super cloud is the reduction of complexity. And I'll explain. He believes anything that takes the requirement of specific expertise out of the equation even a little bit as a CISO worries him. So as you said, they can always two coins, two sides of the coin, but I do believe super cloud is a relevant term and it is necessary because the cloud is continuing to be defined. You know, that's really interesting too. Because Darren, we use snowflake a lot as an example of the early super cloud. And you think from a security standpoint, we've always pushed Amazon. Are you ever going to kind of abstract the complexity away from all these primitives? And their position has always been, look, if we produce these primitives and offer these primitives, we can move as the market moves. When you abstract, then it becomes harder to peel the layers. But Darren, from a data standpoint, like I say, we use snowflake a lot. I think of like Tim Berners-Lee when web 2.0 came out, he said, well, this is what the internet was always supposed to be. So in a way, you know, super cloud is maybe what multi-cloud was supposed to be. But I mean, you think about data sharing, Darren, across clouds, it's always been a challenge. You know, snowflake always, you know, obviously trying to solve that problem as are others. But what are your thoughts on the concept? Yeah, I think the concept fits, right? It is reflective of, it's a paradigm shift, right? It things as a pendulum have swung back and forth between needing to piece together a bunch of different tools that have specific unique use cases and they're best in breed and what they do. And then focusing on the duct tape that holds them all together and all the engineering complexity and skill. It's shifted from that into the pendulum all the way back to let's streamline this, let's simplify it. Maybe we have budget crunches and we need to consolidate tools or eliminate tools. And so then you kind of see this back and forth over time. And with data and analytics, for instance, a lot of organizations were trying to bring the data closer to the business. That's where we saw self-service analytics coming in. And tools like Snowflake, they did was they helped point to different databases. They helped to unify data and organize it in a single place that was in a sense neutral away from a single cloud vendor or a single database and allowed the business to kind of be more flexible in how it brought stuff together and provided it out to the business units. So Snowflake was an example of one of those times where we pulled back from the granular multiple points of the spear back to a simple way to do things. And I think Snowflake has continued to kind of keep that mantle to a degree and then we see other tools trying to do that. But that's all it is. It's a paradigm shift back to this kind of meta abstraction layer that kind of simplifies what is the reality that you need a complex multi-use case multi-region way of doing business. And it sort of reflects the reality of that. And to me, it's a spectrum as part of SuperCloud 2 we're talking to a number of practitioners. Iona's Pharmaceuticals, US West, we got Walmart and it's a spectrum. In some cases, the practitioners saying, the way I solve multi-cloud complexity is monocloud. I just do one cloud. Others like Walmart are saying, hey, we actually are building an abstraction layer ourselves to take advantage of it. So my general question to both of you is, is this a concept that is the lack of standards across clouds, really a problem? Or is SuperCloud a solution looking for a problem? Or do you hear from practitioners that, no, this is really an issue. We have to bring together a set of standards to sort of unify our cloud estates. Allow me to answer that at a higher level. And then we're going to hand it over to Dr. Brabham because he is a little bit more detailed on the real-time streaming analytics use cases, which I think is where we're going to get to. But to answer that question, it really depends on the size and the complexity and the business of your business. At the very large enterprise, Dave, yes, 100%. This needs to happen. There is complexity. There is not only complexity in the compute and actually deploying the applications, but the governance and the security around them. But for lower end or business use cases and for smaller businesses, it's a little less necessary. You certainly don't need to have all of these. Some of the things that come into mind from the interviews that Darren and I have done are financial services, if you're doing real-time trading. Anything that has real-time data metrics involved in your transactions is going to be necessary. And another use case that we hear about is in online travel agencies. So I think it is very relevant. The complexity does need to be solved. And I'll allow Darren to explain a little bit more about how that's used from an analytics perspective. Yeah, go for it. Exactly. I mean, I think any modern multinational company that's going to have a footprint in the US and Europe in China or works in different areas like manufacturing where you're probably going to have on-prem instances that will stay on-prem forever for various performance reasons. You have these complicated governance and security and regulatory issues. So inherently, I think large multinational companies or company and or companies that are in certain areas like finance or in online e-commerce or things that need real-time data, they inherently are going to have a very complex environment that's going to need to be managed in some kind of cleaner way. They're looking for one door to open, one pane of glass to look at, one thing to do to manage these multi points. And streaming is a good example of that. I mean, not every organization has a real-time streaming use case and may not ever. But a lot of organizations do. A lot of industries do. And so there's this need to use, they want to use open source tools. They want to use Apache Kafka, for instance. They want to use different mega cloud vendors offerings like Google PubSub or Amazon Kinesis, Firehose. They have all these different pieces they want to use for different use cases at different stages of maturity or proof of concept, you name it. They're going to have to have this complexity. And I think that's why we're seeing this need to have sort of the super cloud concept to juggle all this, to wrangle all of it. Because the reality is it's complex and you have to simplify it somehow. Great, thanks you guys. All right, let's bring up the graphic and take a look at anybody who follows the breaking analysis, which is co-branded with ETR. Cube Insights powered by ETR knows we like to bring data to the table. ETR does amazing survey work. Every quarter, 1200 plus 1500 practitioners that answer a number of questions. The vertical axis here is NetScore, which is ETR's proprietary methodology, which is a measure of spending momentum, spending velocity, and the horizontal axis shows here the overlap, but it's the presence, pervasiveness in the data set, the ends. That table insert on the bottom right shows you how the dots are plotted, the NetScore and then the ends in the survey. And what we've done is we've plotted a bunch of the so-called super cloud suspects. Let's start in the upper right, the cloud platforms. Without these hyperscale clouds, you can't have a super cloud. And as always, Azure and AWS up and to the right, it's amazing, we're talking about 80 plus billion dollar company in AWS. Azure's business is in the, if you just look at the IaaS, is in the 50 billion range with, I mean, it's just amazing to me the NetScore's here. Anything above 40%, we consider highly elevated and you got Azure and you got Snowflake, Databricks, HashiCorp, we'll get to them. And you got AWS right up there at that size. It's quite amazing with really big ends as well, 700 plus ends in the survey. So kind of half the survey actually has these platforms. So my question to you guys is, what are you seeing in terms of cloud adoption within the big three cloud players? I wonder if you could comment, maybe Eric, you could start. Yeah, sure. Now we're talking data. Now I'm happy. So yeah, we'll get into some of it. Right now the January 2023 thesis is approaching 1500 survey respondents. One caveat, it's not closed yet. It will close on Friday, but with an end that big, we are over statistically significant. We also recently did a cloud survey and there's a couple of key points on that I want to get into before we get into individual vendors. What we're seeing here is that annual spend on cloud infrastructure is expected to grow at almost a 70% CAGR over the next three years. The percentage of those workloads for cloud infrastructure are expected to grow over 70% as three years as well. And as you mentioned, Azure and AWS are still dominant. However, we're seeing some share shift spreading around a little bit. Now to get into the individual vendors you mentioned about, yes, Azure is still number one. AWS is number two. What we're seeing, which is incredibly interesting, Cloudflare is number three. It's actually beating GCP. That's the first time we've seen it. What I do want to state is this is on net score only, which is our measure of spending intentions. When you talk about actual pervasion in the enterprise it's not even close. But from a spending velocity intention point of view, Cloudflare is now number three above GCP and even Salesforce is creeping up to be at GCP's level. So what we're seeing here is a continued domination by Azure and AWS. But some of these other players that maybe might fit into your moniker. And I definitely want to talk about Cloudflare more in a bit but I'm going to stop there. But what we're seeing is some of these other players that fit into your super cloud moniker are starting to creep up Dave. Yeah, I just want to clarify. So as you also know, we track IaaS and pass revenue and we try to extract. So AWS reports and its quarterly earnings, it's, you know, they're just IaaS and pass. They don't have a SaaS play in a little bit maybe. Whereas Microsoft and Google include their applications. And so we extract those out. And if you do that, AWS is bigger. But in the surveys, customers, they see cloud, they sass to them as cloud. So that's one of the reasons why you see, you know, Microsoft as larger in pervasion. If you bring up that survey again, Alex, the survey results, you see them further to the right and they have higher spending momentum, which is consistent with what you see in the earnings calls. Now, interesting about Cloudflare, because CEO of Cloudflare, actually Cloudflare itself uses the term super cloud, basically saying, hey, we're building a new type of internet. So what else, what are your thoughts? Do you have additional information on Cloudflare, Eric, that you want to share? I mean, you've seen them pop up. I mean, this is a really interesting company that is pretty forward thinking and vocal about how it's disrupting the industry. Sure, we've been tracking for a long time and even from the disruption of just a traditional CDN where they took down Akamai and what they're doing. But for me, the definition of a true super cloud provider can't just be one instance, you have to have multiple. So it's not just the cloud, it's networking aspect on top of it. It's also security. And to me, Cloudflare is the only one that has all of it, that they actually have the ability to offer all of those things. Whereas you look at some of the other names, they're still piggybacking on the infrastructure or platform as a service of the hyperscalers. Cloudflare does not need to. They actually have the cloud, the networking and the security all themselves. So to me, that lends credibility to their own internal usage of that moniker super cloud. And also again, just what we're seeing right here that their net score is now creeping above a GCP really does state it. And then just one real last thing, one of the other things we do in our surveys is we track adoption and replacement reasoning. And when you look at Cloudflare's adoption rate, which is extremely high, it's based on technical capabilities, the breadth of their feature set. It's also based on what we call the ability to avoid stack alignment. So those are again, really supporting reasons that makes Cloudflare a top candidate for your moniker of super cloud. And they've also announced an object store and a database. So it takes a while as you well know to get database adoption going. But they're ambitious and going forward. All right, let's bring the chart back up. And I want to focus Darren in on the ecosystem now. And really we've identified Snowflake and Databricks. It's always fun to talk about those guys. And there are a number of other data platforms out there, but we use those two as really proxies for leaders. We got a bunch of the backup guys, the data protection folks, Rubrik, Cohesity and Veeam, they're sort of in a cluster. Although Rubrik, you know, ahead of those guys in terms of spending momentum. And then VMware, Tanzu and Red Hat is sort of the cross-cloud platform. But I want to focus Darren on the data piece of it. We're seeing a lot of activity around data sharing, governed data sharing. Databricks is using Delta sharing as their sort of place. Snowflakes is sort of this walled garden like the App Store. What are your thoughts on, you know, in the context of super cloud, cross-cloud capabilities for the data platforms? Yeah, good question. You know, I think Databricks is an interesting player because they sort of have made some interesting moves with their data lake house technology. So they're trying to kind of complicate, they're trying to take away the complications of, you know, the downsides of data warehousing and data lakes and try to find that middle ground where you have the managed, the benefits of a managed govern, you know, data warehouse environment, but you have sort of a lower cost, you know, capability of a data lake. And so, you know, Databricks has become really attractive, especially by data scientists, right? We've been tracking them in the AI machine learning sector for quite some time here at ETR. Attractive for a data scientist because it looks and acts like a lake but can have some managed capabilities like a warehouse. So it's kind of the best of both worlds. So in some ways, I think you've seen sort of a data science driver for the adoption of Databricks that has now become a little bit more mainstream across the business. Snowflake, maybe the other direction, you know, maybe it's a data warehouse, a cloud data warehouse that, you know, is starting to expand its capabilities and add on new things like Streamlit is a good example in the analytics space with apps. So you see these tools started starting to branch and creep out a bit, but they offer that sort of neutrality, right? We heard one IT decision maker we recently interviewed that referred to Snowflake and Databricks as the quote unquote, Switzerland of what they do. And so there's this desirability from an organization to find these tools that can solve the complex multi-headed use case of data and analytics which every business unit needs in different ways and figure out a way to do that in an elegant way that's governed and centrally managed that federated kind of best of both worlds that you get by bringing the data close to the business while having a central governed instance. So these tools are incredibly powerful and I think there's only gonna be room for growth for those two especially. I think they're gonna expand and do different things and maybe join forces with others. And a lot of the power of what they do well is trying to find these connections and find these partnerships with other vendors and try to be seen as the nice add-on to your existing environment that plays nicely with everyone. So I think that's where those tools are going, but they certainly fit this sort of label of trying to be that super cloud neutral layer that unites everything. Yeah, and if you bring the graphic back up, please. There's obviously big data plays in each of the cloud platforms. Microsoft, big database player, AWS is 11, 12, 15 data stores and of course, BigQuery and other data platforms within Google. But I'm not sure these guys, the big cloud guys are going to go hard after so-called super cloud, cross-cloud services. Although we see Oracle getting in bed with Microsoft and Azure with a database service that is cross-cloud, certainly Google with Anthos and you never say never with AWS. I guess what I would say guys, and I'll leave you with this, is that just like all players today are cloud players, I feel like anybody in the business or most companies are going to be so-called super cloud players, in other words, they're going to have a cross-cloud strategy. They're going to try to build connections. If they're coming from on-prem, like a Dell or an HPE, or Pure or many of these other companies that Cohesity is another one, they're going to try to connect to their on-prem estates, of course, and create a consistent experience. It's natural that they're going to have some consistency across clouds. The big question is, what's that spectrum look like? I think in the one hand, you're going to have some, you know, maybe some rudimentary instances of super cloud or maybe they just run on the individual clouds versus where Snowflake and others, and even beyond that are trying to go with a single global instance, basically building out what I would think of as their own cloud, and importantly, their own ecosystem. I'll give you guys the last thought. Maybe you could each give us closing thoughts. Maybe, Darren, you could start, and Eric, you could bring us home. Just this entire topic, the future of cloud and data. Yeah, I mean, I think, you know, two points to make on that is this question of these, I guess what we'll call legacy on-prem players, these mega vendors that have been around a long time have big on-prem footprints, and a lot of people have them for that reason. I think it's foolish to assume that a company, especially a large, mature multinational company that's been around a long time, it's foolish to think that they can just uproot and leave on-premises entirely full scale. There will always, almost always be an on-prem footprint from any company that was not natively born in the cloud after 2010, right? I just don't think that's reasonable anytime soon. I think there are some industries that need on-prem, things like, you know, industrial manufacturing and so on. So I don't think on-prem is going away, and I think vendors that are going to, you know, go very cloud forward, very big on the cloud, if they neglect having at least decent connectors to on-prem legacy vendors, they're going to miss out. So I think that's something that these players need to keep in mind is that they continue to reach back to some of these players that have big footprints on-prem and make sure that those integrations are seamless and work well, or else their customers will always have a multi-cloud or hybrid experience. And then I think a second point here about the future is, you know, we talk about the three big, you know, cloud providers, the Google, Microsoft, AWS as sort of the opposite of or different from this new super cloud paradigm that's emerging. But I want to kind of point out that they will always try to make a play to become that. And I think, you know, we'll certainly see someone like Microsoft trying to expand their licensing and expand how they play in order to become that super cloud provider for folks. So also I don't want to downplay them. I think you're going to see those three big players continue to move and take over what players like cloud vendor or student cloud flare are doing and try to, you know, cut them off before they get too big. So keep an eye on them as well. Great points. I mean, I think you're right. First point, if you're Dell, HPE, Cisco, IBM, your strategy should be to make your on-prem estate as cloud-like as possible and, you know, make those differences as minimal as possible. And, you know, if you're a customer, then the business case is going to be low for you to move off of that. And I think you're right. I think the cloud guys, if this is a real problem, the cloud guys are going to play in there and they're going to make some money at it. Eric, bring us home, please. Yeah, I'm going to revert back to our data and this on the macro side. So to kind of support this concept of a super cloud right now, you know, Dave, you and I know we back overall spending. And what we're seeing right now is total year spend is expected to only be 4.6%. We ended 2022 at 5%, even though it began at almost eight and a half. So this is clearly declining. And in that environment, we're seeing the top two strategies to reduce spend are actually vendor consolidation with 36% of our respondents saying they're actively seeking a way to reduce their number of vendors and consolidate into one. That's obviously supporting a super cloud type of play. Number two is reducing excess cloud resources. So when I look at both of those combined with a drop in overall spending reduction, I think you're on the right thread here, Dave. You know, the overall macro view that we're seeing and the data supports this happening. And if I can real quick, a couple of names we did not touch on that I do think deserve to be in this conversation. One is HashiCorp. HashiCorp is the number one player in our infrastructure sector with a 56% net score. It does multiple things within infrastructure and it is completely agnostic to your environment. And if we're also speaking about something that's just a singular feature, we would look at Rubrik for data backup storage recovery. They're not going to offer you your full cloud or your networking, of course, but if you're looking for your backup recovery and storage, Rubrik also number one in that sector with a 53% net score, two other names that deserve to be in this conversation as we watch it move and evolve. Great, thank you for bringing that up. Yeah, we had both of those guys in the chart and I failed to focus in on HashiCorp and clearly a super cloud enabler. All right, guys, we got to go. Thank you so much for joining us. Appreciate it. Keep this conversation going. Always enjoy talking to you, Dave. Thanks. Yeah, thanks for having us. All right, keep it right there for more content from SuperCloud too. This is Dave Vellante for John Furrier and the entire CUBE team. We'll be right back.