 So it takes everyone, I'll try to speak loudly, I don't want to stop in any time to ask questions, because some of the talk through virtualization of the central office, and having those nice answers, that works, and then I realized I had about half an hour to do so, it's been a long time. So what I wanted to do, how many people here have actually operated in the central office, or worked in the central office? So there's some people who don't know what I'm talking about, probably going to die. I only work there very rarely. But I want to go through what is the central office, as a starting point. So a little bit of a guide tour, then, of the central office, and a little bit of pictures. To start with, this is what the central office was, when we started talking to these agents. It was a bunch of people coming into a room, and there were ladies there, but me and them, they had to take me up the stairs across the connection. We were on in time, we got a little bit more intuitive when we automated this, and then we had these machines, and I don't know if anyone knows, or if they've ever heard of the sound. If you look up the sound of the central office, right here, it was just a wondrous bore, and it was connected, it was all mechanical, it was big stacks, they were just coming in, switching the thing around, and connecting both of them. Moving around the world, we got a digital, and then we started connecting to the tables, and then you went up to the situation, coming into the central office from somewhere, and then you had very simple computers in the front, then connecting up, and doing what the digital devices used to do. This is what tells me that it became affordable for people, to be able to scale these things up, and you had massive buildings in the middle cities that had downloads and downloads of tables coming in. And then we moved into a digital age, and we moved into being able to multitask, and we were able to put a lot more onto one table. And the central office, and you were all excited, and it started to look something like this. And this is how the central office looks in a small one. Today, you still get large, large central offices in the major metropolitan areas, but for the most part, the access please to the central office is pretty much something like this. It's a real business for a building, and it has a budget for quickly. And today, if you walk into the central office, there's something like this. There's a guide, there's a bunch of different boxes. And each of these boxes, then it comes with productivity, whatever the physical connection, it has to come in, come in to it. It has an operating system, generally built by whoever built the box. Based on what we've done so far today, they still try to really convey the hardware is built for purpose, the software is built for purpose, and you basically wrap these things in a spectrum of all that you share up with your family and people, and you actually have a terminating influence on the central office. So today, it's from the optical connections and from the optical connections, depending on where the central office is. So we always have, today, the central office. Back in the 80s, I think it was, the staffs probably seen the data center, because they started to build computers, the quality of the computers. But today, they're a little bit complicated still. It's too complicated. Each of these boxes has been placed that someone has been able to work with, someone has to be able to interact with, someone has to be able to manage the problem. I may have a bunch of these boxes from Eretton, I might have gone away, I might have just gone all the way to the center, in all the different managing interfaces. You talk to some large telecommunications operators, and they literally have 400 different management interfaces, they have to work with, and they have thousands of people who are specialists on how to work with one of these boxes. So they have problems trying to directly manage their costs, because if you want for us to make phone calls, they have to maintain the thousands of people who manage these things. For most people, for the most part, we don't have a company team. That's another set of people, another set of people. So, the central office today is a publicly-advised, it has a lot of purpose-built equipment that serves telecommunications fields. What we want the central office to look like is it should be beautiful, it should be simple. It should be the place that Trump wants to hang out if he likes them, and not build it. So, really, what we're trying to do is remove the speciality, get real new input of the current, which is that very good, great solution, and move towards a beautiful dark center, beautiful hardware area, and a system platform to manage these interfaces. Then I'm running the physical integration on top of that. So that's the target, that's where we want to go. The central office, the two red pictures, that's more or less in five minutes, now we're all in the same place, we don't want to talk about that, I mean, we don't want to get into it. Then we get the technology, I guess. From a technology perspective, I have my central office on the far side, and the central office is containing some of the examples that provide the rates are out of, B&G, the enemy, signal gateway, aggregated way, for connecting your house up, connecting your phone up. Yeah, so really, let's switch off the first slide. Very briefly, we're all in the same place. I'll let Charles try to sort that one out. My picture's not coming up very well. But in general, you'll have in your house a DSLU, something that connects up to the central office, it's not required to be Wi-Fi, you may have made a plug-in, which connects to all people cable coming into your house. Generally, DSLU is where you'll come from, it doesn't actually play with the wall. The card is not for connection, so it's for the wall. Or maybe you're just using the supporting network of your computer and radio in your house using Wi-Fi. Yeah. Most of the ways of getting access essentially this way to the internet that will provide access to it is to the cable and it connects to your house. There is a mobile service network that connects you up. Mobile service network is interesting. There's a random network, it's a radio access network that's built in genetically dimensioned power lines to the radio so they can speak to your phones and devices. That's separate for the most part to the DSL network that connects your mobile that connects your Wi-Fi on the internet which in this case is totally separate from the on-network that they determine from something in your house. Each network uses different technologies, each network generally uses practically a slightly different way and each network then terminates into the dark side of the way. So we still don't have to consist of where you're getting that into the central office where these boxes are which is why they don't have to go into the dark side of the room and the central office still needs like this. There's still going to be actually not for these version of boxes. But how can we force them? The idea and what we're talking about and what we're talking about is the access and virtualizing to the office is to give them jobs. We want to standardize infrastructure. We want to consist of sensors. We want the same compute. We want to standardize infrastructure isn't compute, right? That's how we do dark centers and the other dark centers. Usually they're going to be made also normal. That's a dark center. Then we have the network. How do they come in? How do they talk to tables? How do they also come in? How is it being managed? How is it isolated or or the different protocols and congratulations that come in? And we also enable easily self-reployable. So if you have a common time barrier and you have hopefully a simple network how do they build software and problems with this remembering we're back to slide. These boxes usually are quite a few years in the making. A lot of engineering goes into this. It's not just put it together and it works. There is a good purpose. It's optimized to run and the best that can possibly run. There is a lot of effort between the engineer and the worker, Disney. The challenge is so I'm an operator and I find this box and I can manage a hundred million users and the throughput is what I don't want to happen is I don't want that to fail. So what I'm going to do is I'm going to run a random point of capacity because in peak hour maybe I need to go up to 80% of the capacity and if I go over 80% of the capacity the software starts to fail. So generally what's happening with these things is if you're running a high-end capacity there's going to be a lot of capacity there's going to be a lot of high-end capacity. So what we want to do is get the software out there in a more flexible manner like that so we want to be able to put software and it should scale and we're going to leave and we're not using software to be able to move the way so it's not going to fail. It's going to open source. I'll get there. Unfortunately the simple office is a physical thing and what we want to do is to understand I guess where we're trying to go is really important. When we talk about the future of the central office we talk about a couple of things. We talk about the point aggregation that we're working on which theoretically as far as we can go Fiber to the something Fiber to the closest point to the consumer that we get is generally the way we do this. So if I move this way out to the greater base station please go out to upon devices and if you go out to the point to the United Front or to the space that you're going to go out there will be some sort of point aggregation that will be standard. Okay. But that doesn't really solve the problem because I still have a network out there and I still have a dark center here and I need to be able to bridge this from the things that we have that's all that we can be able to do. I think this is how I started and what that talks about is essentially creating a new device that terminates all the T in the dark center. So this connects as hard for them to see it. There is a small secret to the connections over here and that provides you to be connected into that box and then a coordinated network going out to the edge directly into and directly to the network in the dark center. What that means is that all of a sudden I need to be aware of the network in the dark center. But we start to talk about virtualizing the CO we start to terminate the OOT actually into the dark center for itself not by the network on its own. And we start to talk about virtualizing the CO and we start to terminate the OOT actually into the dark center for itself and you can see it that's what we're trying to wait for. So we look at what do that's in front of there and then we see how does this create this environment and what's missed in order that I can run a virtual provider or a virtual network and then terminate the OOT and actually direct you over wherever the OOT function. What does CO mean? For those that aren't aware of it it's who in our staff is also making it essentially integrates carrier technologies open source carrier technologies into the stack so you can go into the network stack and that works fine but you can go into the network in a v-stack and that will come with the analyzer in the HSEIO was reporting the U.S.F.C. they would be they would allow us to deploy into a part of the environment like the central office so I'll take come on go I'll take three four minutes to talk about the OOT it's kind of important from every perspective one of the focuses on what we do is we try to bring in new components in the network and stack environment we try to make sure that we're not specializing in short we have consistency basically increasing across the stack usually in the analyzer there's a bunch of projects in the analyzer that come in in a couple later and then for that plane you could be running on an R-cars you could be running on an Intel they say 6,000 seconds they don't like you say that so you can run ODP or you can run FDIO or you can run OBS before hanging on what you want to achieve or what you want to do with the structure you're running and then we run a bunch of tests on that so one of the things that we do is around that 8 hours of testing when you deploy it so it's a full CIC solution it will deploy it after it's on demand and then it will run a bunch of tests stronger than they like stronger than the stack mostly it'll run virtual IMS and then in a sense it'll come out but that's not quite where we need to be yet some of the key things that come out in virtual CIS is what BLT is of course but it changes the fabric and the direction why the data is involved so BLT is the key piece of where we need to go we require fabric access quality so you can't just plug these in because they work you need to be actually isolated so we do require fabric access it's not STMT that we need for it's more STMT that we want a bunch of network projects that require that service or a lot of projects that require that service or orchestration system like a load also require VM to hang up and make it software and life cycle management to be able to deploy direction melt on the workplace so a couple of areas in the open source community which is highly progressive Onos has spawned a project called CORE and CORE is the central office reimagining the data center and they are focusing exclusively on this problem bringing BLT from the project Onos and CORE focus mostly on overflow based control so they assume each other to work in with them very much to bring the work to how I would remember the daylight Overflow has a network it has 3S and 3S and it has the first CORE project these three things the first a network project allows you to network and stack and provide to the 2S to create AS stuff allows you to get to the community VMs and containers and a totally different solution VMs and VMs Workloads is a key part to extending that network activity and all the virtual CO which is the fabric management that that's kind of a hard part that's the part where I come to be able to have the activity of the devices and that's where some of these projects and open loads which are wherever it is there's a project management and orchestration it's also probably a different way that it looks down to the network sees the the capabilities that have to make a network that has to make a network that has to make a network that has to make a network that there's a function which is that has a each轉 just gives a lot of experiment each合 intersection Answer theIs Nevertheless I had all And you saw us, what's the point, it's going to be important without this age, and it's essential office, it's going to be building up the country, and I would really care. Fair enough, what is the point, it's just essential office. The point is, it's the land it had for us to develop a community. It provides immediate access to the community. We have to have this type of universal solution that's going to pass to the community and talk to the local data center that the access to this type of universal solution changes dramatically to directly get access to assets from the provider. It provides global flexibility and employment flexibility. Let's say I want you to send all this and let's say I have a forms of power solution. Then if I have something like this, the universal solution that I have that I have before the control, I can set aside the virtual network layers to access roadside rooms for instance, maybe have a bunch of them running on the roadside, and this type of lights, they can be crafted until it's workable. I should be able to deploy workloads to the balance groups from my self-employed office directly and have more of a very flexible way of deploying the applications and it's right, it's no longer running in the cloud and that's the point. I can work with my job where I need to be able to. And essentially it provides to the data center and the house connected perfectly with the IIT, for instance, where our house is a little bit more open. In this instance, I have, as I said, all the connections from my house to the data center. The IIT services that I can place is running that data center from my house and from my presentation as well. I'm going to assign the way we can see the way this work is going to be done. I'm going to pass the container. Instead of this wire press, push the board. Essentially it reduces the complexity by no more than 8,400 different dimensions. No more than 8,000 people are going to have a problem. We have to move the system where it didn't access the work going so that you know where it's expected. They have to reduce the amount of technical culprits that have to keep the road with the house to be managed and operated together. If they want to run a new education, they've got to run the road and stuff. If they don't want to do that, well, they already know that that is going to be perfect. That's an experience for them. So I guess the point that I really want to give you a talk about is there is a change in how we're doing access networking. It is going to be taken through VLT offices to change how we're doing that work and what's going to change. But it all derives from these communities. I don't really know which of these communities you're interested in. At the end of the day, you're going to have access to it. You're going to have more flexibility and you're going to have multiple costs. I didn't have to pay AT&T to sustain 3,000, 4,000 different money or anything like that. You're going to have to charge you less and charge you less. That's all good. And it's all based on open spaces. Most of this is coming from open-source communities solving specific problems. We need to be able to get a picture. The open system, this is going to be a great role for the weekend. And this is going to be one of these projects that will help you achieve your goals today. And I have a little spreadsheet of science on kind of how to apply, which is good. I just wanted to come today and give you an overview, an introduction. We're going to hear from a number of projects today which are participating in the open-source system, I guess. Maybe they're thinking about those awards, which will see you as a candidate and how far it's going to be in the data center and how far it's going to be. So with that in mind, I'm going to put my presentation here. I don't even have a microphone, so I'm going to repeat the question. Thanks. Yeah. So what about you? It looks to me like you might just probably solve this problem as well by a software-based solution. Is that correct? So at least in the central office, the OOT will be replaced more or less by software. So we don't have software. We still have devices out there managing the terminating. You lower OOTs on the whole site and so forth. But what's happening, I guess, in that domain is that you start to get people through the white holes-type solution. It isn't physical. You might want to have software like on top. At the OOT site, you start to see people think, well, if I'm going to have a device terminating there and an OOT for a workload loader, I'm going to add some instruments, and you see tomorrow I'm going to add smart processes to that. And that's with real time, I think, it's a better start to the point. Can I just take some... I'm having more questions. Do you want to set it up? I'll tell you about these questions and we can start on that. Also, can we decompress to the T? We try to have people in the stairwell. Can we have some more? 3 hours to the stairwell soon. 3 hours to the stairwell soon. All right. So let me say goodbye to Cheryl. Bye.