 Hi. Good afternoon. Welcome to our research deep dive on containers. My name is John Garrett, the founder and partner of ClearPass Strategies. We are an international political research and strategy company. We work for heads of state political parties all around the world from Ukraine to Georgia, Australia, Austria to get this out of the way. No, I don't work in the US. No, I don't think Donald Trump will win. And no, you can't blame me if he does. In addition to our political work, I've been working with Sam Ramji at Cloud Foundry Foundation since 2008, back when he was an open source strategy lead at Microsoft. He brought me and the company I was with at the time in to help with what he saw as more of a campaign related challenge with selling open source from the Microsoft perspective or becoming more open. And so when he went to the Cloud Foundry Foundation, he brought me in to help with another campaign like project, a little project called Cloud Foundry, and how Cloud Foundry can be brought to compete against the likes of AWS, Microsoft, Google, et cetera. So here I am, political consultant talking about containers. We do regular surveys for Cloud Foundry Foundation, broad IT pros and executives. This is what we've done our four surveys and the one we're going to talk about today is the fourth iteration, the latest research today. In July, based on the third round of research, we did a report on containers called Hope vs. Reality, which is based on a survey that was exclusively on containers. And this is an update on that research at the time, how six months later the industry and the use of containers has evolved. To help me do this, someone actually into the tech sector. The real star of the show, many of you know her already, Abby Kearns. Abby is an 18-year veteran of the tech sector, currently is the VP of industry strategy at Cloud Foundry, Cloud Foundry Foundation, and she's been part of the CF community, the Cloud Foundry community for a couple of years now. She's decidedly unencumbered by career in politics and bullshitting like myself, so you can trust her, maybe not so much me. What you won't find in her bio online, she is an Olympic triathlete and a circus lion tamer. Abby, so why don't you tell us a little bit about what you've been doing with Cloud Foundry and then we'll kick it off. I don't know where to go with the lion tamer. So before we get started also, a little bit about this research, this will all be released in a report over the next couple of months, but this is a sneak peek, so you are seeing what no one else has seen to date. We do want this to be an interactive session, so if you have questions, please feel free to ask them. Otherwise, we're going to try to end a little early to offer time at the end for you to ask questions about the data, or if you want to drill down a little bit deeper. The research we released that he referenced in June around hope versus reality was really interesting to us around containers, container usage. Having been in TAC a long time, I anecdotally thought I understood the market. Turns out I did not, and it's amazing how data can really unveil usage patterns as well as trends, and I think that's what's going to be the really interesting part of the conversation today. Great. Thanks. So let's get started. And before we get started with the data, I wanted to start with a little cartoon, someone from my office sent to me, which I thought was really appropriate, and it's a Dilbert cartoon. And to me, this mindset, this mentality sums up what a lot of enterprise things was, whether it's DevOps or containers, you know, rub a little bit of containers on it and poof, all of a sudden digital transformation. That is not how it actually works, is it, Abby? No. Containers will not transform you any faster. But we did learn through this data, and as well as the data back in June, is that more and more people are using containers, and they are using them for workloads that aren't necessarily density-centric. So one of the key points that we made in our report in June was really that, while containers have been around, I mean, for most of you here, you're like, containers aren't new. Containers have been around since 1979. The use of containers really picked up in 2005 with Solaris zones. But containers were used historically for improving density and increasing density in the existing hardware you have. But what we saw in June was a shift in that trend. So using containers less for increasing and improving density, and more for shifting the way that you work with application workloads. So using them more along the lines of versioned run times and consistent build environments, and leveraging them as part of CICD, using them for continuous delivery type solutions. So in short, no, it will not magically make you better, but it is a great first step. Right. So it's a part of the landscape, the tech landscape that everyone's evaluating as they move towards CICD, as they move towards or down the journey towards digital transformation. But it's not the solution by itself. No, but it helps you frame that up. Is you're thinking about digital transformation and realistically, here at this summit, when we talk about digital transformation, we're talking about digital transformation alongside cloud and the use and adoption of cloud, both public and private. And as we think about the value of containers and containerized workloads, it allows you to think about the cloud and think about the abstraction of your application as you move towards those types of workloads. Great. Okay. So finally, let's start getting towards numbers. First things first, I'm a pollster, which means I have to have my hygiene slide up here. This is the methodology slide. This is what we did. Hi, Devin. This is who we talk to. It's a broad IT Pro and executive survey. As I mentioned, it's not a cloud foundry survey or a user survey. It's not a developer survey. It's a broad IT Pro and an exec survey. It's global in nature. We talked to folks in eight different countries, translated into the native language of those countries. It's not just enterprise. About two-thirds of it are enterprise. It's not just enterprise. It's across the roles, devs, ops, managers, line of business. So it's a broad look. It's also a more informed audience. We like listening to respondents that actually know what they're talking about. So some people, when they're taking online services, just click, click, click and go right through. We make sure we have some traps in there to make sure people are actually informed and not just speeding through the survey. So if there's any built-in inherent bias in the survey, it's that they're actually maybe a little more informed than the broader audience. All right, so let's dive in. So the first slide. And just so you know, all numbers are percentages, unless I say otherwise. So as you see here on the left-hand side, that's our previous survey, 47%. What this is showing, this is showing container deployment. Who's using containers, companies that are using containers, the percentage of companies using containers, and at what stage? So you see on the left-hand side, six months ago, 47% of companies were not using containers, which means 53% were, and that declines. That's the major trend on this slide. The other trend on this slide is that the usage that we're seeing increase, 22% to 35% of container usage, it's all in the early stages. So POC, trial stage, that's where the use, that's where the increase is. What we're not seeing is any uptake in the limited deployment, rolling out in business groups or the broad deployment, rolling out across the entire company. And so as even though container usage is increasing, what we're seeing here in this data, what the data is showing me here, is that the depth and breadth of the usage hasn't increased. That stayed flat. Why is that? Well, even more interesting, we referenced earlier the research we did in June, was that most of the container usage was early stage. It was POCs, it was under evaluation. With most of those saying that within the next six to 12 months they would be moving to broad deployment. So we termed it mainstream in that research, but essentially moving these containerized workloads from dev and test and moving them out into production. And so we were expecting when we did this research to see a massive upswing in that. And turns out we were wrong. Because what ended up happening is, yes, more people are using containers, but they're still using them in POC and development type environments. Which means they haven't quite taken that leap to take things from internal dev and test and moving that to production. And while we're seeing more people using it, which is great, if you'll notice the percentages, we're still sub 50. While we talk a lot about containers and containers have been part of our conversation for a couple of years within this community, we're still not seeing broad adoption. And what's really interesting for me, particularly being intact and being in the community, we talked about microservices, we talked about containers for several years now. But looking at how people are actually using these in organizations around the world, they're not there yet. And so we use a lot of buzzwords, microservices, containers, cloud native application architectures. But most people are still trying to figure that out and how to make use of those technologies. So six months ago, what you're saying is that the users were saying that in six to 12 months we were going to be mainstreaming containers. We'd be using containers crossing to our company. Clearly that hasn't panned out. So they were all wrong. But how about you? What do you think? Six to 12 months from now? Or when are containers, when are we going to see broader adoption of containers? Or is it just a fad? They're going to play with it and they're going to kick it out. Well, don't think containers are a fad, nor do I think the cloud is a fad. Even Larry Ellison's bought into it. So clearly it's not a fad anymore. I would say the next six to 12 months we'll start seeing broader adoption. I think some of the other data that we're going to talk about here in terms of workloads and use cases, we're starting to see an uptick in cloud-native application architectures. We're starting to see an uptick in use and then the conversation within our community with large organizations that are starting to work towards digital transformation, which means they're trying to figure out how to take cloud and use cloud and make that part of their technology-centric approach. So I think we'll see, starting to see an uptick. I don't think it's going to be a massive swing in the next six to 12 months, but I think we'll start seeing both an increase in container usage as well as an increase of those broad deployments. Great. And you alluded to cloud-native architecture. Conveniently, this is our next slide. How are people using containers? And what the data is showing us on the left hand are the workloads that they have deployed in containers on the right. We'll get to the right in a second. And what we're seeing there is what the data is showing basically, a rough split between microservice applications and legacy Java applications, maybe leaning more so towards cloud-native applications, if you include the 12 factor on the bottom and the Node.js as well. So it's kind of both, but more leaning towards cloud-native. Is that a correct read of this data? Yeah. And I love this slide because I think we, I personally made a lot of assumptions about what people were using containers for. I think we, depending on the community that you're in and the organization you're in, you make a lot of broad assumptions around what people might use containers for. But it's interesting to see the breakdown with cloud, as you said, cloud-native application architecture is taking the lead. While nearly half of them are still using them for legacy Java applications, we're still majority looking at cloud-native application architectures. And that's encouraging to me, going back to my prediction going forward, is we're starting to see that shift where more people are using cloud-native application architectures in containers, particularly with microservices, which means more people are taking advantage of the capability and the technology. But they're still also, especially with the larger organizations, trying to figure out what to do with the legacy applications that they have. Can I move those to the cloud? Or can I put those in the container? Can I wrap them in APIs and make them accessibility via services, but building out to microservices to take advantage of those and still be able to take advantage of all the things containers and cloud can give to me? Right. So containers are fundamental to this transition. As companies move towards more cloud-native architectures, also still trying to deal with their legacy Java apps, containers are fundamental to that transition. Well, containers are fundamental to take an advantage of cloud. When we think about what are the things you want when you want to move into the cloud, you want elasticity, you want scale, you want the ability to have flexibility with your workload. And containers allow you to have that with your application. You can move it, you can use it, you can scale it out, you can scale it up, you can take advantage of the cloud in that way. And I don't know that I'm going to set up here and advocate that by wrapping your existing legacy application or monolithic application in a container and putting it on the cloud is going to magically provide you cloud-native properties. But I do think it is an important step towards understanding value of container. And if you look at what some of these organizations and people are using containers for, when I come back to standardized deployments and standardized environments, I think, okay, well, this is a great first step in building out continuous delivery practices and a great first step in really understanding the value of cloud towards your application workloads. And what about the right-hand side of this graph? Those are the languages people are using in their containers. That's a lot of Java. It is a lot of Java. What does that mean? What does that tell us? How does that fit in with the story you're telling? It's a lot of Java out there. I think there is a lot of Java, although I was excited to see 40% C++. So it looks like there's a tremendous amount of complex workloads going into containers, which is exciting. But we're still looking at majority Java. And I think as we look at the majority of cloud native application architectures, that's the workloads, with the majority of the language use cases being Java, we can assume that Java is primarily being used to develop these microservices in cloud native application architectures. And I think we've seen that in some of the other data within some of the other communities with the uptake and spring and node and other communities spring and like the uptake and spring source and the ability of it and how it's being used with microservices in the cloud. And I think this really reinforces that use case. Great. And so I'll move to the next slide. And you had mentioned earlier about how larger enterprises are taking advantage of containers and using them to help to make use of their legacy Java apps. And what this slide shows the data here shows a nice progression. The larger companies, reinforcing what you just said, tend to be more focused on the legacy Java apps, the smaller companies, either because they don't have them and they're starting from scratch, tend to be more in the cloud native architectures. Is that progression fit with what you're seeing and hearing? Does it reinforce the notion of companies either straddling the two worlds or moving away from one world to the other world? How do you see this data broken down by enterprise? Well, it's not larger. It's giant. Giant. Excuse me. Giant enterprise. To use your nomenclature. The giant enterprises. I don't think any of you would be surprised with the fact that those organizations that are running legacy Java applications in containers are predominantly giant enterprises. I don't think anyone would be surprised by that data. I do think it's interesting that still 34% of those giant enterprises are working with microservice applications, which going back to where we started in terms of predictions and predicting the future, I think I would wager that the next six to 12 months we'll start seeing that number increase. And either the legacy either stagnating or decreasing as they're making more use of microservice applications to take advantage of legacy data. I hate to use the word bimodal IT, but realistically, if you are a large enterprise or a giant enterprise, you are not going to ditch your legacy environment and move directly to cloud native tomorrow. There is going to be a shift. Just like digital transformation for super enterprises will take a long time. And digital transformation takes five to seven years for large organizations. And so I think we'll start, we'll see that straddle for some time, but it's amazing to see how containers can help bridge that. They can provide that scale. They can provide that that entrance into continuous delivery practices as you try to become a more software-centric organization. Great. Thank you. And thank you for jokes at my expense. I always appreciate that. The final slide here, this one strikes me as super interesting. So maybe we saved the best for last, especially interesting for anyone not buried under a rock who hasn't missed all the buzz about Kubernetes and MISOs, which has been what all the news has been about. When I look at this slide, the data here, what we're seeing on the left is data showing how people say they are managing their containers broadly. And what we're seeing is near a majority, 49% are using either self-managed or provider-managed platform as a service to manage their containers with a much smaller share, only 11% now, saying they're using self-managed orchestration tools, Kubernetes and MISOs and the like. And on the right-hand side, we're seeing pretty strong uptake of Kubernetes, if you include Chorus Tectonic, within the orchestration tools, but the orchestration tools are actually still a small space, and mostly people are managing their containers using platform as a service. In politics, I live in a world where perception is reality that clearly is not the case here because the perception is all about Kubernetes, Kubernetes, Kubernetes, that's not what the data seems to show. So anecdotally, you were talking earlier, anecdotally containers are everywhere, but maybe they're not, anecdotally Kubernetes is everywhere, maybe it's not. How many of you in here have heard of Kubernetes? How many of you have heard of MISOs? How many of you are actually using it? Either. Right. So the point of this and the point of that exercise was to really articulate that it has captured a lot of market buzz, these orchestration tools. There's been a proliferation of them and they have escalated, particularly over the last six months. We've seen Kubernetes scale and grow in terms of brand awareness and marketing and messaging as well as outreach, which I will be the first to say that it's an amazing tool as well as MISOs. But there was this perception that they have taken over and were gaining in mass adoption. But when we look at the people that are using containers and we look at people that are using containers and are trying to use containers at scale, that penetration within the market was not there. And that's important not to say that I'm undermining the value of Kubernetes and MISOs because I'm not because I think they're great technologies in their own right. But when we look at organizations that are contemplating running containers at scale and they're looking at running containerized workloads in large, particularly large and giant enterprises that have other needs and other aspects that they have to be worried about like security, compliance, health monitoring, health management, routing, identity management, all of these things that they have to worry about, they start looking to platforms to help them manage and manage that scale and manage that breadth of their organizations and maintain the availability needs that they require. Great, thank you for that. Now we only have a few minutes left so I thought I'd pause here instead of following up and ask if anyone in the audience has any questions they'd want to ask of Abby or of me, but presumably Abby. If anybody has any questions, open it up please. So the question is how would the graph on the right look if you broke it down by company size including giant enterprise? Well first, I'll pass it off to Abby in a second, but first of all it's a very small sample size. The people on the right that's only 61 people. So breaking that out by company size, we're getting into very qualitative anecdotal level. My hunch is probably less valuable than Abby's hunch in terms of your anecdotal understanding from different companies. Well, Mr. Data tells you that I shouldn't even be talking about it, but what's your anecdotal take about this, Abby, since I can pull up the data but it's very grain assault level data. So anecdotally as you talk because you talk to folks all across the industry, as you talk to them, what is your sense? Where do you see more Kubernetes uptake? My instinct would be these smaller companies, more NIML companies, when we've done focus groups around the world, Beijing or what have you, we get a sense that the enterprise level, they're worried about the maturity level of these companies, so maybe they're focused more on a platform. Is that what you're hearing? Because that's what I'm hearing, but you talk to more industry insiders, so maybe you've got different insights. I feel like, and I've said this in a lot of different panels in a lot of different ways, but essentially I feel like there is, it comes back to the right tool for the right job, and I think there's a place for Mezos, I think there's a place for Kubernetes, and I think there's a place for a platform like Cloud Foundry, and oftentimes all in the same organization, because it's entirely dependent on what you're trying to do and the scale at which you're trying to run. I know anecdotally, there are many large organizations or super organizations that are running all three, and they're running all three in tandem, and they're using them for very different things, and I think we'll continue to see that for some time, because I think we're starting to see them all shake out to offer very different specialties. For Mezos, we've seen a lot of use cases where customers that have needs for state are using that in tandem for our smaller workloads or stateful requirements, and I think for Kubernetes there might be a case for specific protocol requirements, and so I think the technology is all evolving, all three are actually more, because I probably shouldn't forget Docker, all of them are evolving and are evolving at a really rapid rate, and I think the technology, I think about it almost as a line, we have the technology evolving at a rate, but then in parallel to that, we have what we've talked about here with the platforms and the users, the users are evolving at a rate too, so as we're seeing this uptick in usage of containers, and as we're seeing the uptick of use of cloud native application architectures, all of a sudden you're developing more apps, and you're getting more speeding, you're getting more traction, and your needs change, and that continues to evolve, and the ask is going to be different of the technology, and I mean, as we think about where we are today at the end of September, we're so far away from where we were in Q1 of this year and where the technology was and where the users were, and I think we'll continue to see that evolve, and my prediction right now is that we'll see that evolve probably every quarter, if not every six months at an almost exponential rate in terms of the shift and the adoption. Thank you for the question. We have maybe another minute or two, we could probably squeeze in another question if anybody else has a question. You could totally squeeze another two questions in. Another two questions. If not, I have one. Yes? Alex. So the question is, are there people that don't understand, don't realize they're running containers? That is a very fair question. Amazing question. My instinct, because it's almost impossible to ask that question, do you realize that you're using something you're not using? That's not true. The data from earlier this year, that was actually one of the questions. Do you realize you're using something you're not using? Yes. Remind me how we phrase the question. Do you realize using? Do you understand what you're using? Can you define it? Well, I do. We do ask in every survey, so we ask people to define certain terms, do they understand certain terms, and in particular we have a hard time with platform as a service, which we defined at the beginning of every survey, because people have a different definition of platform as a service. Is AWS a platform? Well, not really, but it has platform-like features as dev add-ons, et cetera. And so we know going into this that it's a very fuzzy space already. So we know there's going to be a subset of people in this that either think they are or don't think they are. We asked a question in this survey that's not included here. We asked a battery of questions about is this deployment, whether it's AWS or Azure or Cloud Foundry, is this a platform as a service or is it something else? And you find a large share of people saying, AWS is a pass, and equally large share of people saying AWS is not a pass. In fact, a larger share of saying AWS is not a pass. Point being, you have a broad range of fuzzy range of answers. So my instinct is that, yes, there are a lot of people using containers that don't realize they're using containers and probably some people who think they are that aren't. Anything to add? No, I think you nailed it. I think it was super interesting in the research we did earlier in the year that there were a lot of people that were using a platform, but then when asked to define it, they compared it against things that were not a platform. And I think the same can be true of containers. And it's just going back to where we started, we use a lot of words in our industry. And to quote Duncan Nguyen, you're using that word and I don't think you know what that means. We talk about that a lot because we talk about microservices, containers, cloud native, 12-factor app. We drop those terms all those time. I drop those terms all the time. But then as we reach out to people that are actually using and running these things in organizations and I think they're still struggling with, I don't really understand what that means. And what does that mean in respect to me? Great. Thanks for the question. Thanks, Alex. So we're at two afters. Is there a session in here after us? Are you all coming in for another session after this one? Just trickling in. They're all nodding. So probably, even though no one's kicked us off, we should probably leave. Oh, you're here for this? Oh, you are here. So you're kicking us off. All right. So then that's that. I'm happy to answer additional questions. Anyone has any? Abby, I'm sure, has got a million things going on. Come find us after we're around. But come find us after we're around. Thank you very much for attending.