 We heard GE talk about compliance, we heard Verizon talk about capabilities. Next we're going to have the U.S. Army Cyber School talk about cost. So let's welcome Major Rodriguez and Captain Apsey from the U.S. Army. Good morning, Open SAC. So if you didn't notice, we are from the government, but our specific organization is an educational organization. And we only stood up about two years ago. And when we stood up, we wanted to make sure when we're approaching education that we're approaching it from a perspective that enables us to grow as technology grows, because that's kind of our field with cyber. So our school is now teaching about 500 students annually. And we don't actually choose what we get to teach. It's directed down from above. U.S. CyberCon gives us joint standards, so we're the same as the Air Force, Navy, Coast Guard. And we also have to take orders, of course, from the Army itself. So what they tell us we teach, we teach. How we teach it is up to us. So when we take a look at the end states that we want our students to have, which is around a problem-solving capability, how do we get students to that problem-solving capability? One of the things that we noted, a lot of entry-level IT types courses do is what I like to call breadcrumb learning. So you do breadcrumb exercises where it starts at step one, then you go to step two, you go to step three. Maybe they didn't tell you one part of step four, but basically you know what you're supposed to do there. And at the end of it, okay, now I've got this skill, which applies to this situation, great. The problem is when we're looking at cyber and being able to train soldiers to operate in that domain and solve problems in that domain, from a warfare perspective, we need them to think a little deeper. We need them to be able to come up with their own algorithms. So the way that we wanted to do that was to enable instructors to be able to identify quickly what are the types of problems that we need to put in front of the students. A legacy way of doing this was that the instructor would get an idea, great, this is going to be able to be posed to students in such a way that they're going to learn something awesome out of it. Unfortunately though, the instructor can't create that themselves if we rely on a legacy system where we then write requirements, submit them forward. Hopefully we already have a situation in place that allows us to answer those requirements. But if we don't, maybe we're waiting 12 to 18 months or more to be able to see that idea take place, which doesn't lead to happiness among the students because they're not learning fresh new material and it doesn't lead to happiness in the instructors because the ideas they have, they don't probably see to fruition and then they have to get moved somewhere else by the army. So instead, we're approaching it as an everything as code for courseware, an agile type of solution where you get the idea as an instructor, you immediately go to the computer, you type it in and you submit it through the GitHub Flow and then your students may be using that the next day. So this really rapid pace of being able to speed up an idea to fruition in a problem solving perspective is because we're using OpenStack and other open source software to be able to put together infrastructure as a service. So our infrastructure as a service is named Broadband Handrail. The idea is that the staircase of learning, we wanna make sure that you don't fall off the staircase of learning so we're providing a Broadband Handrail so students can in a semi-controlled environment be able to spin up solutions or answer the solutions for their classes. So as you see in the middle, we have our classes using this but then we also have one of the factors that tie into cost is former students are able to do a reach back use case to access any problems that they did before when they were students in class because as we know in this field if you're not doing it sometimes your skills atrophy but that enables them to then not have to buy their own servers and set up their own kind of solution to be able to do that. So it's both, we save in cost from both the individual soldier perspective and we save in cost overall as an organization because we're using not just OpenStack but also Ceph and SaltStack right now. So these open source projects enable us to avoid these legacy models of only ever going with a highly licensed and costly solution. Okay, so where we started, if you see the picture behind me it's kind of like a LAN party. We had a couple servers in the side of a classroom, they were loud and noisy and it was really just a proof of concept. Is this going to be something that leadership can say, hey, this is worth the risk, let's go ahead and go all in. And it worked. We actually got a lot better feedback from this than a similar solution and this enabled us to get into that flow. Where we're at now, we're still not where we want to be of course and we're always going to be trying to improve and move forward but we do have a couple of standardized racks in a data center which is enabling that student experience a lot happier students over there, no cables hanging from the ceiling and that's giving us the opportunity to look forward and to provide that private cloud solution to our external folks as well as our internal folks. So we're going to show you a little bit of what that looks like. Katnapsi will be giving you a demo. Thanks ma'am. So I'm Katnapsi, I'm the deputy director of the Cyber Technical College at the Army Cyber School. Super cool title for super cool guy obviously. So go ahead and look at this screen. Most of you've probably seen this before, this is a network topology screen within horizon. This is a student demo project, not a haul out gone out right now, we've got the provider network and that's about it. When she mentioned we do everything as code, we really mean that. Infrastructure's code, courseware's code, everything's done with CI pipelines. So we're used to have rooms of people sitting there take away furiously making word documents or PowerPoint slides and whatnot. We got rid of all those people and just do a CI pipeline. So right now, one of our products is the lesson plan which is a rendered PDF based on ASCII docs. ASCII docs is a markup language kind of like markdown. Here's an example that I threw together real quick. It's a lesson network discovery, we have a cool guy quote, everything you need for a good lesson. So to show you how this works, this is a quick sample project. So in every single lesson, there's the course content, which is the PDF you just saw. There's also infrastructure, let's take a look at course content. So this is that ASCII doc markup language I was talking about previously. It's human readable, super easy to use, highly recommend it. Here is that network discovery tag that we saw earlier. Let's go ahead and change that to something else. We'll go ahead and push that up. So this is a live demo. So, all right, so in theory, momentarily we're gonna see some infrastructure pop up in this OpenStack project based on what was contained in that infrastructure folder. Any second now? Just give it a second, you know, it's a pipeline, a lot of very complicated technical things happening on the back end. So just, there we go, all right, cool. So it worked, right? So I just made a quick and dirty change as an instructor and that deployed automatically to the student project, right? That's a huge change from like some model we used to have. We go back down and look at the course content and that changes as well, right? So we used to be network infrastructure, now it says hello OSS. Huge change for us. I know the government hasn't thought a lot as like an innovator, but we like to think that we're kind of moving the ball forward here just a little bit. That's all we have for today. But again, please come see us today at 340 in Ballroom A, we'll have a breakout session, get more into depth about what we do. Thanks very much. Thank you.