 Hi, I'm Brian Langston. I'm a product manager with Morantis. And the things I am a product manager for include Morantis's operation support and tenant care solutions. I've spent the last two years prior to my product management role as a senior cloud architect and the practice lead for North America team. So during that time, we've had a lot of experience in seeing what's really worked across different cloud environments and what some of the good and the bad and the ugly has looked like across different cloud deployments. So what I wanted to focus on for the time today is really what some of those best practices are, what some of the things are that should be avoided in order to improve or increase adoption in private cloud. And this is focused on tenant care specifically. So first of all, as we looked across different clouds, I kind of want to summarize something that we heard from Jonathan Brice on Monday's keynote, and that is that there's this first generation idea, and we're kind of transitioning into the second generation. So using that kind of classification, a couple of things I want to point out here that I might put into this first generation syndrome bucket includes things like a poor or non-existent onboarding experience, right? Many accounts start off with maybe a POC, some kind of prototype, some kind of experimental deployment, then go straight from that into production without really thinking about how am I onboarding tenants and what's their experience like. Other things include maybe a poor cloud design. We've seen some of the more ugly examples include too many services co-located on too few control modes, which obviously leads to issues with instability and inflexibility and difficulty in separating that after you're already in a production mode. Other things, when you can look through the list, maybe some of these resonate with you. But the result in all of this, and this is what we kind of want to highlight as things to avoid, is when our tenants experience many of these things, they start to build up a data sheet of reasons why this is not working for them, and they leave, right? They leave for other alternatives. They leave for public cloud or whatever. So that's something that we obviously want to avoid. And what we should be doing is obviously focusing on delighting our customers, right? Thinking about their experience in our minds. And some of the things that I'm going to talk a little bit more about include things like the smooth and seamless onboarding experience, proactive support. I don't have a Tesla, but for a Tesla owner that I know, I've heard that when there is an issue with Tesla and Tesla knows about it, they proactively call you and say, there is something that's going on. It's OK, bring it in and we'll get it fixed. But that was without the customer even knowing that there was a problem, right? So trying to kind of transfer that same idea, that approach to customer service and tenant care into the cloud experience, I think, can do to wonders. Visibility to workloads, I'll talk a little bit more about this, but some of these other things about clear and open change management, right? I mean, are we really communicating changes that our users ought to be aware of? So the idea here, take care of tenants and your tenants will take care of you. Obviously, if we're in the business of providing a private cloud experience, we have to focus on their experience. And is it positive? Is it something that they are going to stay with? And why? Are they going to tell their peers in your organization or whoever your customers are that it's been a positive experience? So what can we do? In Mirantis, we've been working on various tenant care solutions. I think the way I want to frame this is really about the cloud design, right? The platform services, what are those services that you're designing for? What kind of experience are you intending to deliver? So that starts with the design. Once you've got a design in place, really, it's all about, all right, how do I onboard? How do I get off of maybe my bare metal legacy enterprise IT environments into the cloud? Once tenants are there, what do they have? Do they have any visibility? Do they have access? Do they have awareness of what those resources are? So we'll get into a little bit more detail here in the next couple of slides. So platform services, again, starts with cloud design. Maybe some questions that you might want to ask yourself is, is my cloud design to be functional? Does it provide that good user experience? If it started out as a POC mode and you're transitioning into production, how consumable really are these services? Are you really transferring your V1 automation or lack thereof into a production experience? If so, it's not going to pay very good dividends. Am I even offering services that align with what my tenants need? Do I effectively communicate services, features, and status? So is there a role, really, that is focused on really communicating, identifying the customers, working with them to identify what they're looking for? Workload management. Again, once you've got a cloud designed properly for sustainability, reliability, and a good user experience, how do you get those users onto the cloud? We've seen some cases where there is no awareness of really who the tenants are. This might have been born by the first generation syndrome I mentioned earlier, maybe that Jonathan Bryce was referring to in his keynote. When we start from, say, a POC, some kind of prototype experimental mode, do we even know who our tenants are, right? Did we just create capacity and say, hey, we want you to use it, right? Let's start using it so that I can tell my boss that we've done it, right? We have people using it. But as that happens and it evolves into production, you kind of lose the feel, the awareness, of who's out there, right? Who is actually using my cloud? By name, by email address, by workload. So we're trying to get to encouraging that kind of tenant in cloud discovery. And again, who is the role that's working on identifying these kinds of the tenant awareness in cloud discovery? Once we identify who really our tenants are, obviously we have to focus on what the workload is. How do we take the awareness of what they want in their workload and translate that, map that specifically, and put our fingers on actual cloud configurations and features? Once that assessment is done and there are various methodologies for doing that, of course, like I said earlier, maybe your workload onboarding constitutes transitioning some legacy application from an enterprise legacy data center onto the cloud for the first time. Or if you already have a cloud up, maybe you're doing some workload mobility from a cloud to cloud. So of course, there's different tools. And how that's done is downtime tolerated. If it's not, are you going to handhold? Are you going to allow, kind of, provide a self service or kind of self migration opportunity? So what I mean by workload management is really, from environment to environment, this could obviously differ. But so workload management might take different forms across any of those three activities. Our definition of workload management traditionally kind of stops at that point before application lifecycle management. What we typically see is most companies already have a pretty good idea of how to do the application LCM if you're doing application performance management. You've probably already got an APM kind of a tool. So the challenge really is what happens before that. If we were introducing a new cloud architecture, transitioning from VMware to OpenStack, there's the issues that happen all the way up into the point that the workload is on the cloud that needs to be solved. And then typically, again, the application teams have some tooling that can be applied on top of that. So that's workload management. Platform visibility. When I talk about lack of visibility, there's a couple things I want to point out. Conflicting views of workload state. I don't know how many times you guys in your clouds might have had users that call up and they say, I've got an issue with my workload. Something's not working right. And the operator on the other side is going, well, it looks like my APIs are up. My storage is working. My network is working. I don't see your problem. And the phone call lands, roughly. Generalizing, of course. But even that literal example happens too frequently from what we've seen. Where we have a tenant seeing one view of the cloud and the operator seeing a different one, that's not good. That kind of establishes uncertainty, some doubt in the quality and reliability of the cloud. It kind of breeds and fosters distrust. If I can't trust the operators that they're not looking, they're not looking out for me. They're looking out for the platform, the infrastructure. That's dissatisfaction. And again, if tenants develop enough of a rap sheet against their experience, they are going to alternatives. So in our view, platform visibility services should provide better insights to tenants. What can you provide them to give them more visibility to their workloads? Do they have a consistent view of the cloud? Can we kind of close the gap between operators and tenants? There, of course, are other stakeholders in the cloud, technical account managers, cloud business owners. There should be a coordinated view of the cloud across all those key stakeholders. Wrapping up the whole idea here of tenant care, of course, includes program management. Is there somebody who is focused with their role in the organization focused on providing a positive experience for tenants? This could take the form of a customer success manager, a technical account manager, maybe there's some other names for that kind of a role. Is there a feedback loop with tenants? If there is a feedback loop, does an NPS score, is that included, for some organizations that do have an NPS score, is it positive? There are plenty of organizations we've seen that have an NPS score implemented, some methodology to collect that, and it's negative. So are you above zero? So that's kind of the summary of what I wanted to share. Of course, this is a high level overview. Happy to discuss more details offline, but in summary, again, how is the cloud designed? How is it built? Is it built for how many nines availability and resiliency, workload management? How are you getting that your tenants on boarded? And then once they are there, what are, if you're providing infrastructure as a service, are you following that up with some kind of visibility to those tenants? And then, of course, ensuring that you have executive support in the form of a program manager who can focus on the people, the processes, the methodologies, and the technology that are required to make this happen. So thank you for your time.