 From around the globe, it's theCUBE with digital coverage of AnsibleFest 2020. Brought to you by Red Hat. Hello everyone, welcome back to theCUBE's virtual coverage of AnsibleFest 2020. I'm John Furrier, host of theCUBE. We're here virtual, we're not face to face, obviously because of COVID. So we're doing a virtual event, AnsibleFest coverage. We're at Mary Johnston Turner, Research Vice President of Cloud Management at IDC International Data Corp. Mary, great to see you. Thanks for coming on for AnsibleFest 2020. Thanks for inviting me. So obviously cloud management, everything's cloud native now. We're seeing that at VMworld. We got re-invent coming up. Azure's got growth. The enterprises have gotten some religion on cloud native. COVID certainly is forcing that. What are you seeing from your research at IDC around the convergence of cloud strategies? What's the data tell you? What's the research show? Well, obviously with COVID, a lot of folks have pivoted or accelerated their move to the cloud in many ways. And I think what's happening is that we're seeing many, many organizations recognizing they continue to have need for on-prem resources. They're building out Edge. They've got remote work from home. They've got traditional VM workloads. They've got modern cloud native container-based workloads running on-prem and in public clouds and public cloud services. So it's really kind of a striking world of connected clouds is how I'm talking about it increasingly. And I think what that means from an operational perspective is that it's getting more and more challenging for organizations to maintain consistent configurations, stable APIs, security, compliance and conformance. And they're really starting to look at automation as the way to deal with the increasing scale and velocity of change because that's one of the things that's happening. And I think COVID accelerated it is we've seen organizations stand up applications. They never thought they were gonna have to stand up and they not only stood them up very quickly, but then they continue to update them with great frequency often, multiple times a day or a week. And the infrastructure has had to pivot and the workloads have had to migrate. So it's really been a very challenging time for many organizations. And I think those that are coping the best with it are the ones who have been investing in automation, particularly automation in a CICD pipeline and code-based environment. Yeah, you know, you're seeing the releases obviously automation has helped on the agile side. VMs and containers have been a great way to automate how are customers looking at this? Because it seems to be automation is like the first step towards everything as a service, right? So it's, you know, X-A-A-S as it says, as it's called in the industry, services is ultimately the holy grail and all this because you get when automation and services, you know, it used to be automation, automation, automation and now you're hearing as a service, as a service, as a service is the top three priority. So it seems to be a trajectory. How are customers getting, first of all, do you agree with that? And then how do customers think about this? Because sometimes, you know, we're ahead of the customers. Automation is the first step. What's your take on this? And what are customers planning when it comes to automation? Are they thinking as a service? What are you, what are you hearing from the customers? Let's talk a little bit about what we mean by as a service because that's a really interesting concept, right? And I've been hearing this conversation with folks. As a service started, you know, what a decade or more ago, taking things that particularly software that ran on-prem infrastructure or software and put it into shared data centers where we could run multi-tenant environments, we could scale it. And each cloud provider basically got that scale by investing in their own set of infrastructure automations. So whether it was Azure or VMware or whoever, they, you know, they build a whole repeatable, scalable environment that they could control. What's happening now is that we're seeing these control planes get stretched back to on-prem resources. And I think what's really happening is, is that the line about, well, where does the thing physically have to run becomes more of a discussion around the physics of the matter, you know, latency, data volumes, transaction processing, cost of installed equipment. And every organization is making its own choice about what's the right mix in terms of where physically do things have to run and how they wanna manage them. But I think that we're starting to see a abstraction layer coming in between that. And a lot of that abstraction is automation that's portable, that can be applied across these environments, and that can be used to standardize configurations, to maintain standard APIs, to deploy at very fast speed and consistency across all these different resources. And so automation and the related management layer, to me, is that new abstraction layer that actually is gonna allow most enterprises to stop worrying quite so much about what kind of as a service of my buying and focus more on the economics and the performance and the physics of the infrastructure and then maintain consistency with highly automated, repeatable, programmable style environments that are consistent across all these different platforms. It has a great point, it's great insight, I love that. It's almost as if you can almost visualize the boardroom. We need to change our business model as a service. Go do it, climb that hill, get it done. Like, what are you talking about? We're just trying to manage workloads inside our enterprise and outside as they start looking at the workload aspect of it. It's not trivial to just say it, right? So, and containers has really filled the void here. How are customers and how are people getting started with this initial building block of saying, okay, do we just containerize it? Because that's another hand-waving activity which has a lot of traction. Obviously, you put some containers, it's got some goodness to it. Are many people getting started with solving this problem and what are some of the roadblocks of just managing these workloads inside and outside the enterprise? Well, again, I think, yeah, many organizations are still in the early stages of working with containers. Right now, I think our research shows that maybe 5% to 10% of applications have been containerized. And that's a mix of lift and shift of traditional workloads as well as net new cloud native. Over the next couple of years, the most enterprises tell us they think a third of their workloads could be containerized. So, it's ramping very, very quickly. Again, I think that the goal for many organizations is certainly containers allow for faster development, better support of microservices, but increasingly it's also about portability. I talk to many organizations and say, yeah, one of the reasons I'm moving even traditional workloads into containers is so that I have that flexibility. And again, they're trying to get away from the tight coupling of workloads to physical resources and saying, yeah, I'm going to make those choices but they might change over time or I might need to, you know, COVID happens. I have to scale much faster than I ever thought. I'm never going to be able to do that in my own data center. I'm going to go to the cloud. So, I think that we're seeing increasing investments in Kubernetes and containers to promote more rapid scaling and increased business agility. And again, I think that means that organizations are looking for those workloads to run across a whole set of environments, geographies, physical locations, edge. And so they're investing in platforms and again, automation to help them do that. So your point here is that, you know, five, 10% that's a lot of growth opportunity. So containers is actually happening now. So, you know, you're starting to see that progression. So that's great insight. So I got to ask you on the COVID impact, that's certainly changed some orientation because, hey, this project, let's double down on this, is a tailwind for us, work from home, this new environment. And these projects, maybe we want to wait on those. How do we come out of COVID? Some people have been saying some spending in some areas are increasing, some are not. How are customers spending money on infrastructure with COVID impact? What are you seeing from the numbers? Well, that's a great question. You know, and I can see one of the major things we do is track IT markets and spending and purchasing around the world. And as you might expect, if you go back to the early part of the year, there was a very rapid shift to cloud, particularly to support work from home. And obviously there was a lot of investment in, you know, virtual desktops and remote work kinds of collaboration very early on. But now that we're sort of maturing a little bit and moving into, you know, more of a, you know, ongoing recovery resiliency sort of phase, we continue to see very strong spending on cloud. I think overall it's accelerated this move to more connected environments. Many of the new initiatives are being built and deployed in cloud environments. And, but again, we're not seeing a whole, whole hog, you know, exit from on-prem resources. The other thing is edge. We're seeing a lot of growth on edge. Both again, there's sort of work from home, but also, you know, more remote monitoring, more support for, you know, all kinds of IoT and remote work environments, whether it's, you know, lab testing or, you know, data analysis or contact tracing. I mean, there's just so many different use cases. I'm going to ask you about Ansible and Red Hat. I'll see you've been following Ansible since the acquisition by Red Hat. How do you think they're doing vis-a-vis the market? They're competitors that have also been acquired. What's your take on their performance, their transition, their transformation? Well, you know, this infrastructure is code or automation is code. Market has really matured a lot over the last 10 or more years. And I think the Ansible acquisition was about five years ago now. You know, I think we've moved from just focusing on trying to build elegant automation languages, which certainly was an early initiative. You know, Ansible offered one of the earlier human-readable, Python-based approaches as opposed to, you know, more challenging programming languages that some of the earlier solutions had. But I think what's been really interesting to me over the last couple of years with Red Hat is just what a great job they've done in promoting the community and building out that ecosystem. Because at the end of the day, the value of any of these infrastructure's code solutions is how much they promote the connectivity across networks, clouds, servers, security, and do that in a consistent, scalable way. And I think that, you know, that's what really is going to matter going forward. And that's probably why you've seen a range of acquisitions in this market over the last couple of years, is that as a standalone entity, it's hard to build those really robust ecosystems and to do the analytics and the curation and the support at large scale. So it kind of makes sense as these things mature that they become, find homes, you know, with larger organizations that can put all that value around it. That's a great commentary on the infrastructure's code. Totally agree. You can't go wrong by building abstraction layers and making things more agile. I want to get your take on some announcements that are going on here and get your thoughts on your perspective. Obviously, the release of the private automation hub and a bunch of other great stuff. I mean, bringing automation, Kubernetes and a series of new features to the platform together. Obviously, continuation of their mission. But one of the things when I talked to the engineers is I said, what's the top three things at Ansible Fest? And they go, collections, collections, collections. So you start to see this movement around collections and the platform. The other theme is, you know, it's a tool market and everyone's got tools, we need a platform. So it's the classic tools. We saw that in big data in other areas where you start getting into platform, you need management, you need orchestration, you need automation services. What's your perspective on these announcements? They've been investing aggressively. What does it mean? What you're taking? What does it mean? Yeah, I would agree that Red Hat has continued to invest very aggressively in Ansible over the last few years. What's really interesting is if you go back a couple of years, we had Ansible Engine, which included, you know, periodic, maybe every quarter or even longer than that, distributions that pretty much all Ansible code got shipped on. And then we had Tower, which provided an API and a way to do some audit and logging and integration with source control. And that was great, but it didn't move fast enough. And we just got done talking about how everything's accelerated and everything's now connected clouds. And I think a lot of what Red Hat's done is really approach the architecture for scale and the ecosystem for scale. And so the collections have been really important because they provide a framework to not only validate and curate content, but also to help customers navigate it and kind of quickly find the best content for their use cases. And also for the partners to engage, there's a lot, I think it's 50 plus collections now that are focused on partner content. And so I think it's really provided an environment where the ecosystem can grow, where customers can get the support that they need. And then with the Automation Hub and the ability to support really robust source control and distribution, and again, it's promoting this idea of an automation environment that can scale not only within a data center but really across these connected environments. Great, great stuff. I want to get your thoughts because I want to define and understand what Red Hat and Ansible, when they talk about curated content, which includes support for open shifts, you know, versus pulling content from the community. I hear content. I'm like, oh, content. Is that a video? Is that, what is content? So can you explain what they mean when they say they're currently building out aggressively building out curated content? This idea of, what does content mean? Is it content? Is it code? Sure, and yeah, I think, you know, any of these automation is code environments. You really have a set of build blocks that in the Ansible framework would be modules and playbooks and roles. And those are relatively small, stable pieces of code. Much of it is actually written by third parties or folks in the community to do a very specific task. And then what the Ansible platform is really great at is integrating those modules and playbooks and roles to create much more robust automations and to give folks a starting point and an ability to do, rather than having to code everything from scratch to really kind of pull together things that have been validated, have been tested, get security updates when they need it, that kind of thing. And so the customers can focus on essentially chaining these things together and customizing them for their own environment as opposed to having to write all the code from step one. So content means what in this context? What does content mean for them? It's automation building blocks. It's code, it's small amounts of code that do very specific things. And in a collections environment, it's tagged, it's tested, it's supported. And so if there's- Not a research report or a theCUBE video, it's real, it's like code, it's not content, it's real. Right, I know. But again, this is automation as code, right? So it's pieces of code that rather than needing an expert who understands everything about how a particular device or system works, you've got reusable pieces of code that can be integrated together, customized and run on a repeatable, scalable basis. And if they need to be updated because of any API changes or something, there's a chain that goes back to the vendors who again are part of the ecosystem and then there's a validation and testing so that by the time it goes back into the collections, the customers can have some confidence that when they pull it down, it's not going to break their whole environment. Whereas in a pure community-supported model, the content submitted by the community may be beautiful, but you don't know. And you can have five submissions that kind of do the same thing. How do you know what's going to work and what's going to be stable? So it's a lot of helping organizations, get automation faster in a more stable environment. We can certainly follow up on this trend because one of the things I've been digging into is this idea of open source and contribution. Integrations are huge. The collections to me is super important because when you start thinking about integration, that's one of cloud-needest supposable strengths is to be horizontally scalable, integrate, build these abstraction layers, as you had pointed out. So I got to ask you with respect to open source, I was just talking with a bunch of founders yesterday here in Silicon Valley around as cloud scales and certainly seeing snowflake build on top of AWS. I mean, that's an amazing success story. You're starting to see these new innovations where the cloud-scale providers are providing great value propositions and the role open source is trying to keep pace. And so I got to ask you still open source, I mean, I believe it's important, but how does open source maintain its relevance as cloud scale goes on because that's going to force automation to go faster. Okay. And you got the major cloud vendors promoting their own cloud platforms. Yet you got the innovation of startups and companies. Your enterprises are starting to act like startups. As containers starts to get through this lift and shift phase, you'll see innovation coming from enterprises as well as startups. So you started to see this notion of real value on top of these clouds. What's your take on all of this? Well, I think open source and the communities continue to be very, very important, particularly at the infrastructure layer because to get all this innovation that you're talking about, you act, if you believe you've got a connected environment where folks are going to have different footprints and probably more than one public cloud set of resources, it's only going to, the value is only going to be delivered if the workloads are portable, they're stable, they can be integrated, they can be secured. So I think that the open source communities continue to be incredibly important as a way to get industry alignment and shared innovation on the platform and infrastructure and operational levels. And I think that that's going to be something that we're going to see for a long time. Well, Mira, I really appreciate your insights. Got one final question, but I'll just give you a plug for the folks watching. Check out Mary's work at IDC. Really cutting edge and super important is cloud management really is at the heart of all the, whether it's multi-cloud, on-premise hybrid or full cloud lift and shift or cloud native, management plays a huge important role right now. That's where the action is. You look at the container growth, as Mary you pointed out, it's great. So I have to ask you, what comes next? What do you think management will do relative to cloud management as it evolves in these priority environments around cloud, around on-premise as the operations start to move along, containers are critical. You talked about the growth, it's only five, 10%. A lot of headroom there. How is management going to evolve? Well, again, I think a lot of it is going to be as everything has to move faster. And that means that automation actually becomes more and more important, but we're going to have to move from automation at human speed to automation at container and cloud speed. And that means a lot of it is going to have to be driven by AI and ML analytics that can, an observability solution. So I think that that's going to be the next wave is taking these very diverse sources of log and metrics and application traces and performance and end user experience and all these different things that tell us how is the application actually running and how is the infrastructure behaving? And then putting together an analytics and automation layer that can be a very autonomous. We've at IDC been doing a lot of research on the future of digital infrastructure. And this is a really fundamental ten of what we believe is that autonomous operations is the future for cloud and IT. Final point for our friends out there and your friends out there watching who somewhere on the cutting edge riding the big wave of cloud native, they're at KubeCon, they're digging in, they're in service matches, Kubernetes containers, you name it. And for the folks who have just been kind of grinding it out in IT operations, holding down the fort, running the networks, running all the apps. What advice do you give the IT skillset friends out there that are watching what should they be doing? What's your advice to them, Mary? Well, you know, we're going to continue to see the convergence of a virtualized and container based infrastructure operations. So I think anyone out there that is in those sorts of roles really needs to be getting comfortable with programmatic code driven automation and figuring out how to think about operations from more of a policy and scalability point of view. Increasingly, if you believe what I just said about the role of analytics driving automation, it's going to have to be based on something, right? There are going to have to be rules, there's going to have to be policies, there's going to have to be configuration standards. And so kind of making that shift to not thinking so much about the one off lovingly handcrafted environment, the thing about how do we scale, how do we program it and starting to get comfort with some of these tools, like an Ansible, which is designed to be pretty accessible by folks with a large range of skillsets. It's human readable, it's Python based, you don't have to be a computer science major that'd be able to get started with it. So I think that that's what many folks have to do is start to think about expanding their skillsets to operate at even a greater scale and speed. Mary, thanks so much for your time. Mary, Johnston Turner, Vice President of Research for Cloud Management at IDC for the Ansible Fest, virtual, I'm John Furrier with theCUBE, for CUBE coverage, CUBE virtual coverage of Ansible Fest 2020 virtual. Thanks for watching.