 Thanks for joining us and be sure to look up there other sessions. Take care, Jonathan. Thank you. Before our next speaker, we have a short video from Morantis, so let's roll the video. Does Morantis know how to build open-stack clouds? Do Russians know how to drink vodka? Morantis, we know how to build open-stack clouds. Ladies and gentlemen, please welcome Boris Rensky. Thank you, Bear. So, my marketing team asked me to integrate the Bear into the presentation, so I couldn't think of anything other than just put the Bear there in the title slide, so Bear. And now I'm going to dive right into it. So, very dramatic, dramatic change in sentiment from Bear to Piecharts. So, this is a Gartner study. It's kind of like a Gartner-themed conference, it seems like. So, we didn't like Gartner very much in Morantis three years ago when they were saying that open-stack is not ready and open-stack is crap. Now, it seems like they changed their mind. Open-stack seems to be not crap. So, with that, we like Gartner-Morantis. So, whoa. So, we're going to talk about this. So, this is a study that Gartner came out with in May of last year. Problems encountered by 95% of private clouds. And Gartner used a lot of labels to describe the categories of problems, and a lot of his labels are kind of ambiguous. Me being Russian and all, it's kind of hard to figure them out. So, I try to really bring them down to simple terms and dissect them down. So, let's just kind of roll in there. So, the first one we're going to talk about is, this is 31%, by the way, the failures of all private clouds is the failure to change the operational model. So, what does that exactly mean? So, I've kind of tried to translate it, and I haven't practiced internally with some of my colleagues explaining, and it seems to have resonated. So, I'm going to try on you. So, to explain that, I think it's good to look at the public and private side-by-side and just look at the cloud stack. So, at the base of every cloud, there is typically some sort of cloud software. It can be OpenStack, it can be something else, but cloud software has to be there. Now, this is not the end of it. The next thing that's always there is actually the operations team. And oftentimes, people implementing a cloud don't think about it very much. They just think I'm just going to put the software there and it's going to run itself. Unfortunately, that's not the case. Software needs to be run by people. It needs to be patched. It needs to be upgraded. And, ultimately, that's what defines the experience for the cloud tenant. And cloud tenants are there on top. So, when we talk about the failure to change the operational model in the enterprise as the failure for private clouds, what, in my understanding, Gartner really means is that basically when it comes to public clouds, there's a whole bunch of DevOps ninjas that are constantly running around and putting out all kinds of fires so tenants don't notice. But if you go to private cloud paradigm, you get into the enterprise world. And in the enterprise world, what you get is basically a bunch of VMware sysadmins that are trying to tackle the same exact problem. And this paradigm is new to them. So, no matter how committed they are and how hard they're trying, they oftentimes fail, tenants don't get a very good experience, and people blame the software. The point here is that no matter how good the software is, it can never make up for the expertise and competence of the operations team in the organization. Let's look at the next one. This is the second biggest one, and that's doing too little. So, what's doing too little? Very vague. And I didn't understand it myself for a long time, but then I kind of related it to a real story, a real experience that we had at Mirantis. Some years ago, when we were just getting started in OpenStack and trying to adopt an OpenStack cloud internally. So, here's the story. We have a training team that does a whole bunch of training certifications, and they needed a cloud to host the training labs. And they're part of my team, so they reached out to me and I at the same time was lucky to learn that actually our IT has adopted an OpenStack cloud. So, I'm like, okay, well, let's put these guys together, they'll figure it out. So, the first thing that happens is our IT sends a link to the training folks, and the link is so long, it's an IP address and slash and a bunch of stuff, so long that you have to copy paste it into the browser address bar to even get there, impossible to type in. And then when you click enter, you basically get to this. So, the training team looked at it and said, okay, well, how do I log in? Like, isn't it supposed to be integrated with LDAP or what's the deal here? So, they're asking IT, can I get access? And what they get in return is, well, you know, you have to file a Jira ticket and we'll create a tenant for you. Like, okay, well, that's something. So, the next thing is, you know, they open up the Jira page, it says, well, you have to specify the number of VMs you'll be using at peak and how long you need the tenant for. And they're like, we don't know. So, the reply from IT is like, how do we know how to charge to your budget the resource you're going to be using in cloud? And the reply back is, well, you know, it's cloud, right? It's supposed to be on demand, it's supposed to be metered. Isn't that the point? Like, well, you know, actually, yes, but at this first stage of implementation, we just decided to put this thing there and see how people start using it. And if people start using it a lot, we'll do the second stage and we'll add all the cool bells and whistles such as metering. Like, okay, well, great, let me know when it's done. I'm going to go to public cloud for now. So, the point, again, is that open stack that IT could have implemented there. This cluster could have been super solid, excellent HA, never go down. But with this form factor of implementation, there's literally zero value above and beyond just requesting VMs from IT via filing jury tickets. And if you actually go down the circle more and more, you'll see that only 6% of all of the private cloud problems are really attributed to actual technology. Everything else is either people or process. There's nebulous terms describing each one, but it boils down to people and process. So, where does this leave us? Successive open stack is one part technology and nine parts people in process. Another way to think about it is if you're trying to succeed with open stack in your organization and you're embracing open stack just as technology, then you will most likely fail. The interesting thing is that this may be very obvious, but it's amazing how common we start a project at Marentus. And nowadays we try to tackle the people in process problem. And we look for free things. We look for an enthusiastic sysadmin that wants to make open stack succeed. We look for an anchor tenant that we can work with right away so that it can partner. And then we try to define some sort of metrics of, you know, what the success would look like for that tenant, right? So, we almost always get the enthusiastic sysadmin. But the project starts off with, you know, I want open stack with Sahara and Cloudera Hadoop and Contrail Networking and also some Kubernetes maybe Meso sprinkled on the side. The problem with this approach is that this may be a way to go, but the objective of the project is not really necessarily to get the organization to transform. It's really to get the guy that you're working with, which is typically just the enthusiastic sysadmin, to get a bunch of notches on his resume that he worked with all of his cool technologies, where after he can really just kind of go and use his resume to go find another job. And it's a very hard problem to tackle because people in process is boring. Technology is really exciting. So, how do we tackle this people in process problem today in most cases, right? So, there's two ways that you can tackle it and there's two ways that people look at things in terms of extremes really. One way is that you can go private cloud and today most of the time that implies basically taking all of your VMware sysadmins and turning them into DevOps engines. Also changing a whole bunch of process doing all of it yourself and that's really hard. The other option is to go public cloud. And people say that public cloud technology is so great and it's awesome and it stays up all the time. It's not really. I think the value of a public cloud is more about short-cutting the people in process problem in the organization, not so much the technology thing. And today, we kind of see this world in these two extremes. You can private, do everything yourself, or public, which is basically give up on all of your people, give up on all of your process, close all your data centers and go public. Now, if you look at reality, historically, things are not black and white. Things are very gray in the world. And what we see is the two only viable options today, which is public cloud, which is give up on everything, or private, do everything yourself, are really the extremes of a broader scale, which will get increasingly broader as the market continues to grow. And in between those two extremes, there is a whole bunch of other operational models. Some of them we started to see already now, such as managed private cloud, or managed private cloud on-prem, or different startups doing appliances that they can give you and you can manage it through some sort of interface. And we've seen just a few, few of those emerge today. Now, at Morantis, this in-between is really kind of our bread and butter. The reason why we've been successful, in my view so far, is because we started tackling the people-in-process problem in the private cloud space before we actually started addressing the technology problem. Technology has always been the afterthought for us. So, I wanted to close with a little bit of a public cloud bashing. So, I know that public cloud is really, really hot right now. And, you know, I'm fully expecting you guys will stand up and start throwing tomatoes at me. So, you know, beware I'm going to call the bear and it's going to eff you up if you're going to do. So, but I just wanted to kind of put things in perspective. I think that public cloud is a great thing. I'm a big fan of AWS. I think that AWS really started this huge new disruption in the industry. It showed the world how to build applications against the data center instead of building them against the computer. And if it wasn't for AWS, I don't think any of us would be sitting here. OpenStack wouldn't exist. CloudStack, Eucalyptus, all of these cool things would not exist. But putting things in perspective, you have to look at the bigger market. And AWS is just, you know, kicking butt $10 billion in run rate, but there's dot right there. Now, if you look at, and this is gardener statistics, by the way, data center infrastructure virtualization market, it's $130 billion. And this is the market that AWS is disrupting just with EC2, not with all of the other stuff. And then if you look at all of the other aspects of the market, such as, you know, application, delivery, building, et cetera, all of the IT market that basically is being influenced by this disruption, it's $3.5 trillion. So, the point I'm trying to make is that we're just at the beginning of a disruption. And thinking of the world in terms of just public or private is a very simplistic way to think about it. There's going to be many different operational models, such as managed private or appliances or all kinds of things, that we've only seen a sliver of today that will be fueled and created by this disruption. And all of them, all of these operational models, will likely lend themselves to take advantage of a fabric like OpenStack. So, thank you. Thank you, Boris. Thank you. You know, you never know what Boris is going to say, but you know he is always going to tell you what he really thinks.